JP3716647B2 - Safety promotion device for automobile - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a safety promoting device for a vehicle that alerts a driver of the vehicle to a safety confirmation.
[0002]
[Prior art]
As a conventional safety promoting device for automobiles, for example, there is one described in JP-A-7-108850. This safety promotion device obtains the distance from the obstacle to the host vehicle when an obstacle is detected on the road, and stops with a normal brake or a warning from the relationship between the distance and the host vehicle speed, The vehicle is provided with stopping by automatic braking and three-stage stopping means, and after stopping by warning or automatic braking, an alarm for prompting the driver to confirm safety is issued when the vehicle restarts.
[0003]
[Problems to be solved by the invention]
However, with this conventional safety promotion device, an alarm is issued only when the vehicle restarts after stopping due to the detection of an obstacle, etc., so it is possible to call attention for safety confirmation only in very limited situations. There is a problem that safety cannot be promoted in a driving environment that requires safety confirmation during driving. Moreover, since an alarm is issued regardless of whether or not the driver confirms safety, there is a problem that it is troublesome when the driver confirms safety.
[0004]
Therefore, an object of the present invention is to provide an automobile safety promoting device that can prompt the driver to maintain safety when the driver fails to confirm safety in a driving environment that requires safety confirmation.
[0005]
[Means for Solving the Problems]
The invention according to claim 1 is an image input means for inputting a face image of a driver of a car, and a feature amount detection means for detecting and storing a face feature amount by performing image processing on the face image input by the image input means. Driving environment detection means for detecting the driving environment of the host vehicle; and driving environment determination means for determining whether the driving environment of the host vehicle is an environment that requires safety confirmation based on detection data of the driving environment detection means; A safety confirmation determination means for determining whether or not the driver has confirmed safety based on the detection data of the feature amount detection means when the travel environment determination means determines that the travel environment requires safety confirmation; Safety promoting means for urging the driver to maintain safety based on the determination that the safety confirmation by the safety confirmation determining means is neglected The traveling environment detection means detects a change in the steering angle of the host vehicle and a change in the course of the host vehicle by the operation of the direction indicator, and the traveling environment determination means needs to confirm safety when the direction indicator is operated. The safety confirmation determination means prompts the driver to maintain safety when it is determined that the driver has failed to confirm safety during the period from the start of operation of the direction indicator until the steering angle changes. Instructing the safety promotion means It is characterized by.
[0008]
Claim 2 The invention includes an image input unit that inputs a face image of a driver of a car, a feature amount detection unit that detects and stores a face feature amount by performing image processing on the face image input by the image input unit, and the host vehicle A traveling environment detecting means for detecting the traveling environment of the vehicle, a traveling environment determining means for determining whether the traveling environment of the host vehicle is an environment that requires safety confirmation based on detection data of the traveling environment detecting means, and the traveling A safety confirmation determining means for determining presence or absence of a driver's safety confirmation based on detection data of the feature amount detecting means when the environment determining means determines that the driving environment requires safety confirmation; and the safety confirmation Safety promotion means for encouraging the driver to maintain safety based on the determination that the safety check by the determination means is neglected, and the travel environment detection means is configured to check whether the brake pedal is depressed and the gear position of the transmission The travel environment determination means determines that the vehicle is in the reverse position as a travel environment that requires safety confirmation, and the safety confirmation determination means determines that the gear position of the transmission is in the reverse position. When it is determined that the driver has failed to confirm the safety from when the brake pedal is not depressed, the safety promoting means is instructed to encourage the driver to maintain safety.
[0009]
Claim 3 The invention includes an image input unit that inputs a face image of a driver of a car, a feature amount detection unit that detects and stores a face feature amount by performing image processing on the face image input by the image input unit, and the host vehicle A traveling environment detecting means for detecting the traveling environment of the vehicle, a traveling environment determining means for determining whether the traveling environment of the host vehicle is an environment that requires safety confirmation based on detection data of the traveling environment detecting means, and the traveling A safety confirmation determining means for determining presence or absence of a driver's safety confirmation based on detection data of the feature amount detecting means when the environment determining means determines that the driving environment requires safety confirmation; and the safety confirmation Safety promoting means for encouraging the driver to maintain safety based on the determination that the safety confirmation by the judging means is neglected, and the traveling environment detecting means is configured to detect whether the accelerator pedal is depressed and the gear position of the transmission The travel environment determination means determines that the vehicle is in the reverse position as a travel environment that requires safety confirmation, and the safety confirmation determination means determines that the gear position of the transmission is in the reverse position. When it is determined that the driver has failed to confirm the safety from when the accelerator pedal is depressed, the safety promoting means is instructed to urge the driver to maintain safety.
[0019]
【The invention's effect】
According to the first aspect of the present invention, the image input means inputs the face image of the driver of the car, and the feature amount detection means detects the face feature amount by performing image processing on the face image input by the image input means, and stores it. The traveling environment detection means detects the traveling environment of the host vehicle, and the traveling environment determination means determines whether the traveling environment of the automobile is an environment that requires safety confirmation based on the detection data of the traveling environment detection means. The safety confirmation determining means determines whether or not the driver confirms safety based on the detection data of the feature amount detecting means when the driving environment determining means determines that the driving environment requires safety confirmation. The safety promotion means prompts the driver to maintain safety based on the determination that the safety confirmation judgment means has failed, so if the driver fails to confirm safety in a driving environment where safety confirmation is required, the driver To promote safety Kill.
[0020]
In the invention of claim 1, The travel environment detection means detects a change in the course of the host vehicle by the operation of the direction indicator, and the travel environment determination means determines that the travel environment in which the safety check is necessary when the direction indicator is operated. The safety confirmation determination means determines whether or not the driver has confirmed safety, and if the driver neglects the safety confirmation, the safety promotion means can prompt the driver to maintain safety. Accidents can be prevented in advance when changing the lane, overtaking, or changing the course such as turning left or right at an intersection.
[0021]
In the invention of claim 1, The traveling environment detection means detects a change in the steering angle of the host vehicle, and the safety confirmation determination means determines whether or not the driver has confirmed the safety between the start of operation of the direction indicator and the change in the steering angle. Therefore, when the driver operates the direction indicator to change the course, such as changing lanes, if the driver neglects safety after operating the direction indicator and turning the steering wheel, the driver Immediately after starting to cut, the safety promotion means can prompt the driver to maintain safety in accordance with an instruction from the safety confirmation determination means, and accordingly, an accident at the time of course change such as a lane change can be surely prevented.
[0028]
Claim 2 In this invention, the traveling environment detecting means detects whether or not the gear position of the transmission is in the reverse position, and the traveling environment determining means determines that the vehicle is in the reverse position as a traveling environment that requires safety confirmation. When reversing, the safety confirmation determining means determines the presence or absence of the driver's safety confirmation, and if the driver neglects the safety confirmation, the safety promotion means can urge the driver to keep the safety, and therefore when the vehicle reverses Accidents can be prevented in advance.
[0029]
Claims 2 In the present invention, the traveling environment detecting means detects whether or not the brake pedal is depressed, and the safety confirmation determining means is from when the transmission gear position is in the reverse position to when no depression of the brake pedal is detected. In addition, when it is determined that the driver has neglected the safety check, the safety promotion means is instructed to urge the driver to keep the safety, so when the driver neglects the safety check when moving the vehicle backward, The safety promotion means prompts the driver to maintain safety immediately after the release of the foot brake according to the instruction from the safety confirmation determination means, and it is possible to reliably prevent an accident when the vehicle moves backward.
[0030]
Claim 3 In this invention, the traveling environment detecting means detects whether or not the gear position of the transmission is in the reverse position, and the traveling environment determining means determines that the vehicle is in the reverse position as a traveling environment that requires safety confirmation. When reversing, the safety confirmation determining means determines the presence or absence of the driver's safety confirmation, and if the driver neglects the safety confirmation, the safety promotion means can urge the driver to keep the safety, and therefore when the vehicle reverses Accidents can be prevented in advance. According to the invention of claim 5, the traveling environment detecting means detects whether or not the accelerator pedal is depressed, and the safety confirmation determining means detects whether or not the accelerator pedal is depressed from the time when the transmission gear position is in the reverse position. If it is determined that the driver has failed to confirm the safety, the safety promotion means is instructed to encourage the driver to maintain the safety. In such a case, the safety promotion means prompts the driver to maintain safety immediately after the accelerator pedal is depressed by the instruction of the safety confirmation determination means, so that an accident at the time of reverse of the vehicle can be surely prevented.
[0034]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a functional block diagram showing an embodiment of the present invention. As shown in FIG. 1, this safety promotion device includes image input means CL1, facial feature amount detection means CL2, running environment detection means CL3, running environment judgment means CL4, safety confirmation judgment means CL5, and safety promotion means CL6. ing.
[0035]
The image input means CL1 inputs the face image of the driver of the car. The feature amount detection unit CL2 detects the feature amount of the face by performing image processing on the face image input by the image input unit CL1. The traveling environment detection means CL3 detects the traveling environment of the host vehicle. The traveling environment determination unit CL4 determines whether the traveling environment of the host vehicle is an environment that requires safety confirmation based on the detection data of the traveling environment detection unit CL3.
[0036]
The safety confirmation determination means CL5 determines the presence or absence of the driver's safety confirmation based on the detection data of the feature amount detection means CL2 when the travel environment determination means CL4 determines that the travel environment requires safety confirmation. . The safety promotion means CL6 prompts the driver to maintain safety based on the determination that the safety confirmation by the safety confirmation determination means CL5 is neglected.
[0037]
FIG. 2 is a block diagram showing the configuration of one embodiment of the present invention. As shown in FIG. 2, in the present embodiment, a TV camera 11 is provided as an image input means CL <b> 1 for photographing a driver's face from the front on an instrument panel of an automobile. The face image photographed by the TV camera 11 is converted into image data by the AD converter 12. This image data is stored in the image memory 13 and output from the image memory 13 to the image data calculation circuit 14.
[0038]
The image data calculation circuit 14 determines pixels for each local increase in density in the vertical pixel row of the face from the image data as extraction points, and continuous data in which adjacent extraction points are continued in the horizontal direction of the face. The positions of the eyes and eyebrows are specified based on the above. The feature amount recognition circuit 15 sets a predetermined area including the eye and eyebrow positions specified by the image data calculation circuit 14, counts the number of data of eye and eyebrow data existing in the set predetermined area, Recognize driver's facial features.
[0039]
Therefore, the AD converter 12, the image memory 13, the image data calculation circuit 14, and the feature quantity recognition circuit 15 constitute a face feature quantity detection means CL2. The recognition of the facial feature amount by image processing of the driver's face image will be described in detail later.
[0040]
The travel environment detection circuit 21 as the travel environment detection means CL3 detects the travel environment of the host vehicle by various sensors. The determination circuit 22 determines whether the traveling environment of the host vehicle is an environment that requires left and right safety confirmation based on the data input from the traveling environment detection circuit 21, and is an environment that requires safety confirmation. Is determined based on the data input from the facial feature amount recognition circuit 15. Accordingly, the determination circuit 22 constitutes a travel environment determination unit CL4 and a safety confirmation determination unit CL5. The determination by the determination circuit 22 will be described in detail later.
[0041]
The alarm processing circuit 23 as the safety promotion means CL6 includes an alarm device that issues an alarm to the driver with sound or light, and when the determination circuit 22 determines that the driver has failed to confirm safety, an instruction from the determination circuit 22 To alert the driver.
[0042]
Here, a method for recognizing the feature amount of the driver's face by image processing of the driver's face image will be described. FIG. 3 is a flowchart illustrating image processing of a driver's face image, and FIG. 4 is an explanatory diagram illustrating setting of a pixel row.
[0043]
As shown in FIG. 3, in step S <b> 110, the driver's face image captured by the TV camera 11 is converted by the AD converter 12 into a pixel image in which the density of each pixel is a digital value. In step S120, image data corresponding to an image of one frame is stored in the image memory 13.
[0044]
In step S130, a vertical pixel row of the face is selected from the enormous amount of image data stored in the image memory 13 as shown in FIG. That is, one pixel column is set for every 10 pixels in the horizontal (X) direction so that the pixel columns in the vertical (Y) direction of the face are several tens of columns at equal intervals in the central portion of the screen. Note that 512 pixel columns are set in the horizontal (X) direction, and one pixel column has 480 pixels in the vertical (Y) direction, and the density value of all pixels is calculated for each pixel column. Is done.
[0045]
In step S140, a density value is calculated for each pixel column to specify a pixel (Y coordinate value) that serves as an extraction point. In step S150, the extracted points obtained for each pixel column are connected and grouped in the horizontal (X) direction. In step S160, unnecessary connection data is removed from the grouped connection data, and candidate point data is formed only for the finally required connection data. In step S170, eye and eyebrow candidate point data is selected from a plurality of candidate point data, and the positions of the eye and eyebrow are specified. Accordingly, steps S130 to S170 are executed by the image data calculation circuit 14 in FIG.
[0046]
In step S180, a predetermined area including the positions of the eyes and eyebrows is set, and the number of eye and eyebrow data existing in the predetermined area is counted to recognize the feature amount of the driver's face. Accordingly, step S180 is executed by the facial feature amount recognition circuit 15 in FIG. Steps S140 to S180 will be described in detail below.
[0047]
FIG. 5 is a flowchart showing step S140. 6 is a graph showing the density of the pixel row Xa in FIG. 4, where (a) shows the arithmetic mean density value, and (b) shows the differential value of the arithmetic mean density value. FIG. 7 is a graph showing the density of the pixel row Xb in FIG. 4, where (a) shows the arithmetic mean density value, and (b) shows the differential value of the arithmetic mean density value.
[0048]
As shown in FIG. 5, in step S141, the density values of all the pixels for each pixel column are called a predetermined number of times in order from the left side of the image data shown in FIG. In step S142, the density value of each pixel on the pixel column is arithmetically averaged to obtain an arithmetic average density value obtained by removing high frequency components (noise) from the density value on the pixel column. For example, the arithmetic average density value shown in FIG. 6A is calculated from the pixel column Xa shown in FIG. 4 and the arithmetic average density value shown in FIG. 7A is calculated from the pixel column Xb.
[0049]
In step S143, the arithmetic mean concentration value is differentiated. The arithmetic average concentration value shown in FIG. 6 (a) is the differential value shown in FIG. 6 (b), and the arithmetic average concentration value shown in FIG. 7 (a) is the differential value shown in FIG. 7 (b). .
[0050]
In step S144, first, the Y coordinate value corresponding to the dark peak of the arithmetic mean density value is obtained from the differential value calculated in step S143. That is, the Y coordinate value whose differential value is inverted from negative to positive becomes one Y coordinate value representing each of the eyebrows, eyes, mouth and the like corresponding to the darkness peak. Note that the Y coordinate value at which the differential value is inverted from positive to negative is not employed because it is a peak position of brightness corresponding to an intermediate position such as an eyebrow or eye.
[0051]
For example, the arithmetic mean concentration value shown in FIG. 6A becomes the differential value shown in FIG. 6B, and Y coordinate values p1, p2, p3, p4, and p5 at which the differential value is inverted from negative to positive are extracted. Next, in step S144, Y coordinate values q1, q2, and q3 are extracted such that the peak value of the differential value before each Y coordinate value p1, p2, p3, p4, and p5 is equal to or less than a predetermined threshold value (broken line). Identify as a point. Note that it is possible to avoid facial wrinkles and irregularities from becoming extraction points by discriminating the peak value of the differential value in front.
[0052]
From the arithmetic mean concentration value shown in FIG. 7A, Y coordinate values p1, p2, and p3 are extracted in which the differential values shown in FIG. 7B are inverted from negative to positive. However, in any case, the peak value of the previous differential value does not fall below the threshold value (broken line), and no extraction point is formed.
[0053]
In step S145, it is determined whether or not it is the last pixel column, that is, the rightmost pixel column in FIG. If it is the last pixel column, the extraction points have been specified in all the pixel columns, and the process proceeds to step S150. If it is not the last pixel column, the process returns to step S141 to specify the extraction point from the density value data of the next pixel column.
[0054]
The distribution of the extracted points specified in this way is shown in FIG. As shown in this figure, about 0 to 4 extraction points A1, A2, A3, and A4 are specified for each pixel column. For example, the extraction points A1 and A2 are specified in the pixel column Xc, and the extraction points A1, A2, A3, and A4 are specified in the pixel column Xd.
[0055]
FIG. 9 is a flowchart showing step S150 in FIG. FIG. 10A is a graph showing grouping of extraction points. In step S150, as shown in FIG. 10A, the extraction points specified for each pixel column are connected in the X direction to be grouped.
[0056]
As shown in FIG. 9, in step S151, the extraction point data of the pixel column is called in the order of the pixel column. In step S152, the Y coordinate values are compared for the extraction points of the pixel rows adjacent to each other. In step S153, the difference of the Y coordinate values of the extraction points in the pixel rows adjacent to each other is within a predetermined range (10 pixels in this embodiment). It is determined whether it is in. If the difference between the Y coordinate values of the extraction points of the pixel rows adjacent to each other is 10 pixels or less, the process proceeds to step S154 to form concatenated data. If the difference between the Y coordinate values exceeds 10 pixels, the process proceeds to step S155. No concatenated data is formed.
[0057]
In step S154, concatenated data is formed from extraction points of adjacent pixel columns. The concatenated data is classified by the group number given in order for each generated group and the number of the pixel column to which the extraction point at the left end of each group belongs (corresponding to the X coordinate value). The number N (number of pixel columns) and the Y coordinate value are stored in the memory, and each time the pixel column is processed, the memory content is rewritten and formed.
[0058]
In step S155, it is determined whether or not it is the last pixel column. If it is the last pixel column, the grouping of the extraction point data has been completed for all the pixel columns, and the process proceeds to step S160. If it is not the last pixel column, the process returns to step S151 to group the extraction point data of the next pixel column.
[0059]
FIG. 11 is a flowchart showing step S160 in FIG. FIG. 10B is a graph showing candidate points formed from grouped data of extracted points. In step S160, unnecessary connection data is removed, and candidate point data is formed only for the finally required connection data.
[0060]
As shown in FIG. 11, in step S161, the concatenated data is called from the memory one by one in the order of group numbers. In step S162, the number N of linked data is evaluated. In step S163, it is determined whether the number N of linked data is greater than or equal to a predetermined value (5 in the present embodiment). If the number N of linked data is greater than or equal to a predetermined value, the process proceeds to step S164 to form candidate point data. If the number N of connected data is less than the predetermined value, it is determined as, for example, a nostril or the like, and the process proceeds to step S165 and no candidate point data is formed.
[0061]
In step S164, for each group, a candidate consisting of the average value of the Y coordinate values of the concatenated data and the average value of the X coordinate values of the concatenated data (the average value of the concatenation start pixel column number and the concatenation end pixel column number). Form point data. For example, as shown in FIG. 10A, extraction point groups G1 to G6 corresponding to both the eyebrows and the eyes, the shadow of the nose, and the mouth are left from the extraction point group shown in FIG. However, here, the remaining six groups G1 to G6 are renumbered in order from left to right and from top to bottom. Therefore, each group G1 to G6 of extraction points shown in FIG. 10A is represented by six candidate points K1 to K6 shown in FIG.
[0062]
In step S165, it is determined whether it is the last concatenated data. If it is the last concatenated data, since all the concatenated data has been evaluated, the process proceeds to step S170. If it is not the last concatenated data, the process returns to step S161 to start evaluation of the next concatenated data.
[0063]
In step S170 in FIG. 3, eye and eyebrow candidate point data is selected from a plurality of candidate point data, and the positions of the eye and eyebrow are specified. For example, in the case of six candidate points K1 to K6 shown in FIG. 10B, a central detection zone ZC is set around the candidate points K5 and K6 having large (low) Y coordinate values. Outside the detection zone ZC, left and right detection zones ZL and ZR are set with the candidate points K1 and K3 having small (high) Y coordinate values and the candidate points K2 and K4 as centers.
[0064]
The height relationship between the candidate points K1 to K4 included in the left and right detection zones ZL and ZR and the candidate points K5 and K6 included in the center detection zone ZC, and both candidate points K1 and K1 included in the detection zone ZL. Based on the positional relationship between K3 and the positional relationship between both candidate points K2, K4 included in the detection zone ZR, candidate points K3, K4 corresponding to the eyes and candidate points K1, K2 corresponding to the eyebrows are selected, Identify eye and eyebrow positions.
[0065]
In step S180 in FIG. 3, a predetermined area including the positions of the eyes and eyebrows is set, and the number of data of eye and eyebrow data existing in the predetermined area is counted to recognize the feature amount of the driver's face. To do. Here, step S180 will be described with reference to FIGS.
[0066]
FIG. 12 is an explanatory diagram showing face images of the driver's eyes and eyebrows. In FIG. 12, reference numeral 31 denotes an outer frame of the screen, reference numeral 32 denotes a driver's eyebrow, and reference numeral 33 denotes a driver's eye. FIG. 13 is an explanatory diagram showing grouping of extracted points from those shown in FIG. 12 and setting of a predetermined area based on the grouped points. Note that the grouping of extraction points has already been described, and a description thereof will be omitted.
[0067]
In FIG. 13, reference symbols G <b> 1 and G <b> 2 indicate extraction point groups corresponding to the eyebrows 32, and reference symbols G <b> 3 and G <b> 4 indicate extraction point groups corresponding to the eyes 33. Reference numeral 34 indicates a predetermined area (hereinafter referred to as a window) for determining whether or not the driver has confirmed safety. The size of the window 34 is such that the driver's eyes and eyebrows are in the forward gaze state where the driver is gazing forward, and includes the amount of movement of the eyes and eyebrows due to vibration during travel. Yes.
[0068]
FIG. 14 is an explanatory diagram of a processed image showing a state where the driver looks at the left door mirror. As shown in FIG. 14, when the driver looks at the left door mirror from the forward gazing state, the extraction point groups G1 to G4 corresponding to the eyes 33 and the eyebrows 32 move largely in the horizontal direction, although there is no change in the height direction. Then, both the left groups G2 and G4 are removed from the window 34.
[0069]
FIG. 15 is an explanatory diagram of a processed image showing a state where the driver looks at the right door mirror. As shown in FIG. 15, when the driver looks at the right door mirror from the front gaze state, the right-side groups G1 and G3 of the extraction point groups G1 to G4 corresponding to the eyes 33 and the eyebrows 32 are removed from the window 34.
[0070]
FIG. 16 is an explanatory diagram of a processed image showing a state in which the driver looks obliquely backward. As shown in FIG. 16, when the driver looks diagonally backward from the front gaze state, all the extraction point groups G1 to G4 corresponding to the eyes 33 and the eyebrows 32 are removed from the window 34.
[0071]
Accordingly, the driver is gazing at the front of the vehicle based on the direction in which the candidate points K1 to K4 representing the extracted point groups G1 to G4 move and the number of candidate points K1 to K4 existing in the window 34. It is possible to recognize whether the left side of the vehicle has been checked for safety, whether the right side of the vehicle has been checked for safety, or whether the vehicle has been checked for safety behind the vehicle.
[0072]
FIG. 17 is a flowchart for determining whether or not the driver has confirmed safety. As shown in FIG. 17, in step S1, the number of candidate points K1 to K4 representing the groups G1 to G4 existing in the window 34 is counted as the number of data. In the window 34, there are usually four pieces of data for both eyes and both eyebrows.
[0073]
In step S2, it is determined whether or not the number of data existing in the window 34 has decreased. If the number of data existing in the window 34 has decreased, the process proceeds to step S3, and it is determined that the safety has been confirmed. If the number of data existing in the window 34 remains four, the process proceeds to step S4. It is determined that the safety check has been neglected.
[0074]
As described above, in the present embodiment, the feature amount detection unit CL2 specifies the positions of the eyes 33 and the eyebrows 32 from the driver's face image, and within the window 34 including the specified positions of the eyes 33 and the eyebrows 32. When the number of data in the window 34 is the same as the number of data when the driver gazes forward, the safety confirmation determination means CL5 When it is determined that the safety check is neglected and the number of data in the window 34 is smaller than the number of data when the driver gazes forward, it is determined that the driver has confirmed the safety.
[0075]
For this reason, when the driver remains in the forward gaze state in a driving environment that requires safety confirmation on the left, right, or rear, it can be determined that the driver has failed to confirm safety, and safety confirmation on the left, right, or rear is required. When the driver confirms safety by turning left or right or rearward in the driving environment, it can be determined that the driver has confirmed safety. Therefore, the safety confirmation determination means CL5 determines whether the driver has confirmed safety. It can be determined with certainty.
[0076]
FIG. 18 is an explanatory diagram showing a case where the automobile changes lanes. To change the lane, the driver of the automobile 41 operates the direction indicator at the point A of the road 42 on one side two lanes, confirms the safety on the right rear by looking at the right door mirror, and then touches the handle at the point B. Turn right. In FIG. 18, reference numeral 43 denotes a center line, reference numeral 44 denotes a lane change locus of the automobile, and reference numeral L1 denotes a distance between AB.
[0077]
FIG. 19 is a flowchart executed by the determination circuit when a lane is changed in the present embodiment. However, the traveling environment detection circuit 21 shown in FIG. 2 includes a steering angle sensor that detects the steering angle of the steering wheel and an operation sensor that detects whether or not the direction indicator is operating, and the steering wheel is steered by the steering angle sensor. The angle is detected, and the presence or absence of the operation of the direction indicator is detected by the operation sensor.
[0078]
When the automobile 41 changes lanes, as shown in FIG. 19, the determination circuit 22 of FIG. 2 inputs the steering angle of the steering wheel from the steering angle sensor as a steering angle signal in step S11, and in step S12 The presence / absence of operation of the direction indicator is input from the operation sensor as a direction indicator signal.
[0079]
In step S13, the determination circuit 22 determines whether or not the direction indicator is activated based on the direction indicator signal input in step S12. When the direction indicator is operating, the process proceeds to step S14, and when the direction indicator is not operating, the process returns to step S11.
[0080]
In step S14, the presence or absence of a course changing operation is determined based on the steering angle signal input in step S11. If the rudder angle maintained at the point A changes and there is a course changing operation, the process proceeds to step S15, and if there is no course changing operation, the process returns to step S11.
[0081]
In step S15, whether the driver has confirmed safety based on the data from the facial feature amount recognition circuit 15 shown in FIG. 2 from when the direction indicator starts to operate until the steering wheel is steered. Determine whether or not. If the driver fails to confirm safety after the direction indicator is activated and before the steering wheel is steered, the process proceeds to step S16. If the driver confirms safety, the process returns to step S11.
[0082]
In step S <b> 16, the determination circuit 22 outputs a signal to the alarm processing circuit 23 of FIG. 2, and generates an alarm such as a sound or light that prompts the driver to confirm safety from the alarm processing circuit 23.
[0083]
As described above, when the automobile 41 changes lanes in the present embodiment, the determination circuit 22 first determines whether or not the direction indicator is operating, and when the direction indicator is operating, The presence / absence of a course change operation is determined. If it is determined that a course change operation has been performed, the presence / absence of a driver's safety check is determined. If the driver fails to check the safety, an instruction is given to the alarm processing circuit 23. The alarm processing circuit 23 generates an alarm that prompts the driver to confirm safety. Therefore, in this embodiment, it is possible to prevent an accident when changing lanes.
[0084]
Moreover, in the present embodiment, the determination circuit 22 prompts the driver to confirm safety when the driver neglects safety from when the direction indicator starts to operate until the steering wheel is steered. Since an alarm is generated from the alarm processing circuit 23, if the driver neglects safety when changing lanes, an alarm prompting the driver to confirm safety is generated from the alarm processing circuit 23 immediately after the driver turns the steering wheel. Therefore, accidents when changing lanes can be prevented in advance and reliably.
[0085]
By the way, when changing lanes, in addition to the case where the driver confirms safety in the AB section of FIG. 18, the direction indicator is activated after confirming safety and the steering wheel is turned off immediately after the activation. There is also.
[0086]
In order to cope with this case, the determination circuit 22 determines whether or not the driver has confirmed the safety between the time when the direction indicator is activated and the time when the direction indicator is activated. When the determination circuit 22 determines that the driver has failed to confirm the safety, the alarm processing circuit 23 may issue an alarm to the driver according to an instruction from the determination circuit 22.
[0087]
In this way, when the driver turns off the steering wheel immediately after the operation of the direction indicator while neglecting the safety check before operating the direction indicator, the alarm processing circuit 23 immediately after the start of the operation of the direction indicator. Can issue a warning to the driver and prompt the driver to confirm safety, so the lane change is made immediately after activation of the direction indicator without the driver confirming safety before operating the direction indicator. Accidents can be prevented without fail.
[0088]
In addition, since the determination circuit 22 determines the presence or absence of the driver's safety check by going back a predetermined time from the start of the operation of the direction indicator, the safety check was performed before the driver operated the direction indicator. The warning to the driver in the case can be avoided, and the troublesomeness felt by the driver with respect to the warning to the driver who has confirmed the safety can be eliminated.
[0089]
Note that the flowchart shown in FIG. 19 is not only for the case where the automobile 41 changes lanes, but for example, when the driver overtakes by operating the direction indicator, or when the driver turns right or left at the intersection. Needless to say, the present invention can be applied when changing the course of the vehicle 41 by operating the indicator.
[0090]
FIG. 20 is an explanatory diagram illustrating a case where a vehicle passes through an intersection. The driver of the automobile 41 enters the intersection 63 after confirming safety on the left and right by temporarily stopping the automobile 41a at the stop line 64 before the intersection 63 of the traveling road 61 in order to pass through the intersection 63 without a signal. Pass through the intersection 63. In FIG. 20, reference numeral 62 indicates a road that intersects the traveling road 61, and reference numeral 41 b indicates a car that has passed through the intersection 63.
[0091]
FIG. 21 is a flowchart executed by the determination circuit when passing the intersection in the present embodiment. However, the traveling environment detection circuit 21 shown in FIG. 2 includes a navigation system (hereinafter abbreviated as NAVI) and a vehicle speed sensor that detects the vehicle speed, detects road information of the traveling road 61 using the NAVI, and uses the vehicle speed sensor. The vehicle speed of the host vehicle 41 is detected.
[0092]
When the automobile 41 passes through the intersection, as shown in FIG. 21, the determination circuit 22 of FIG. 2 inputs the information on the traveling road from the NAVI as NAVI information in step S21, and from the vehicle speed sensor in step S22. The vehicle speed of the host vehicle 41 is input as a vehicle speed signal.
[0093]
In step S23, the determination circuit 22 determines whether there is a traffic light at the intersection 63 existing on the traveling road 61 from the host vehicle 41 to a predetermined distance ahead based on the NAVI information input in step S21. If there is no traffic light at the intersection 63, the process proceeds to step S24, and if there is a traffic signal, the process proceeds to step S25.
[0094]
In step S24, based on the NAVI information input in step S21 and the vehicle speed signal input in step S22, it is determined whether or not there is a temporary stop before the intersection 63. If the vehicle is temporarily stopped before the intersection 63, the process proceeds to step S25. If the vehicle is not temporarily stopped before the intersection 63, the process proceeds to step S26, and the alarm processing circuit 23 of FIG.
[0095]
In step S25, when entering the intersection 63 before passing through the intersection 63, it is determined whether or not the driver has confirmed safety based on the data from the facial feature amount recognition circuit 15 shown in FIG. If the driver fails to confirm safety when entering the intersection 63, the process proceeds to step S26. If the driver confirms safety, the process returns to step S21.
[0096]
In step S <b> 26, the determination circuit 22 outputs a signal to the alarm processing circuit 23 and causes the alarm processing circuit 23 to generate an alarm such as a sound or light that prompts the driver to confirm safety at a timing such as the rise of the vehicle speed.
[0097]
Note that the alarm generation timing for prompting the driver to confirm safety is not limited to the vehicle speed rising timing based on the vehicle speed signal input in step S22. For example, the traveling environment detection circuit 21 detects whether or not the brake pedal is depressed by a brake switch, detects whether or not the accelerator pedal is depressed by an accelerator switch, and the determination circuit 22 is after the depression of the foot brake is released. An alarm may be generated from the alarm processing circuit 23 by capturing the timing immediately after the start of depression of the accelerator pedal.
[0098]
As described above, when the automobile 41 passes through the intersection 63 in the present embodiment, the determination circuit 22 ensures that the driver is safe when entering the intersection 63 before passing through the intersection 63 regardless of the presence or absence of a traffic light at the intersection 63. It is determined whether or not confirmation has been performed, and if the driver fails to confirm safety when entering the intersection 63, an alarm is issued to instruct the driver to confirm safety by instructing the alarm processing circuit 23 at a timing such as the rise of the vehicle speed. Generated from the alarm processing circuit 23. Therefore, in this embodiment, an accident when passing through the intersection 63 can be prevented in advance.
[0099]
In addition, when there is no traffic signal at the intersection 63, the determination circuit 22 determines whether or not the vehicle has stopped before the intersection 63. The alarm processing circuit 23 is instructed to generate an alarm from the alarm processing circuit 23 that prompts the driver to confirm safety. Therefore, in this embodiment, it is possible to prevent an accident caused by ignoring stoppage at the intersection 63 where there is no traffic signal.
[0100]
Furthermore, in this embodiment, since the traveling environment detection circuit 21 as the traveling environment detection means CL3 includes the NAVI, it is possible to reliably detect the intersection 63 existing on the traveling road 61 from the host vehicle 41 to a predetermined distance ahead. It is possible to reliably detect whether or not the intersection 63 has a traffic light, and it is also possible to detect the stop position at which the vehicle should stop at the intersection.
[0101]
For this reason, in the present embodiment, it is possible to reliably detect whether or not the own vehicle 41 has stopped at the stop position, and whether or not the driver has confirmed left and right safety when stopping at the stop position. Can also be determined.
[0102]
FIG. 22 is an explanatory diagram showing a case where the automobile passes through the railroad crossing. The driver of the automobile 41 temporarily stops the automobile 41a at the stop line 73 before the level crossing 72 of the traveling road 71 and confirms the left and right safety, and then enters the level crossing 72 and passes through the level crossing 72. In FIG. 22, the symbol L <b> 3 indicates the distance from the automobile 41 to the level crossing 72.
[0103]
FIG. 23 is a flowchart executed by the determination circuit when passing through a crossing in the present embodiment. However, the traveling environment detection circuit 21 shown in FIG. 2 includes a NAVI and a vehicle speed sensor that detects the vehicle speed, detects road information of the traveling road 71 using the NAVI, and detects the vehicle speed of the host vehicle 41 using the vehicle speed sensor. Yes.
[0104]
When the automobile 41 passes the railroad crossing 72, as shown in FIG. 23, the determination circuit 22 shown in FIG. 2 inputs the information of the traveling road 71 from the NAVI as NAVI information in step S31, and in step S32, The vehicle speed of the host vehicle 41 is input from the vehicle speed sensor as a vehicle speed signal.
[0105]
In step S33, the determination circuit 22 determines whether or not there is a railroad crossing 72 from the host vehicle 41 on the traveling road 71 to a predetermined distance ahead based on the NAVI information input in step S31. If there is a level crossing 72, the process proceeds to step S34. If there is no level crossing 72, the process returns to step S31.
[0106]
In step S34, based on the NAVI information input in step S31 and the vehicle speed signal input in step S32, it is determined whether or not there is a temporary stop before the level crossing 72. If it is temporarily stopped before the level crossing 72, the process proceeds to step S35. If it is not temporarily stopped before the level crossing 72, the process proceeds to step S36, and the alarm processing circuit 23 of FIG.
[0107]
In step S35, it is determined whether or not the driver has confirmed safety based on the data from the facial feature amount recognition circuit 15 shown in FIG. To do. If the driver fails to confirm the safety before the vehicle restarts after the temporary stop before the level crossing 72, the process proceeds to step S36, and if the driver confirms the safety, the process returns to step S31.
[0108]
In step S36, the determination circuit 22 outputs a signal to the alarm processing circuit 23 and causes the alarm processing circuit 23 to generate an alarm such as a sound or light that prompts the driver to confirm safety at a timing such as rising of the vehicle speed.
[0109]
As described above, in the present embodiment, when the automobile 41 passes the railroad crossing 72, the determination circuit 22 first determines whether or not the railroad crossing 72 exists in front of the predetermined distance from the own vehicle 41 on the traveling road 71. If there is a level crossing 72, it is determined whether or not there is a temporary stop before the level crossing 72. If it is temporarily stopped before the level crossing 72, the driver confirms the safety before restarting. If the driver fails to confirm the safety, the alarm processing circuit 23 is instructed to issue an alarm from the alarm processing circuit 23 to prompt the driver to confirm the safety at a timing such as the rise of the vehicle speed. . Therefore, in the present embodiment, an accident at the time of passing through the level crossing 72 can be prevented without fail.
[0110]
Moreover, if the determination circuit 22 has not stopped once before the level crossing 72, the alarm processing circuit 23 generates an alarm for instructing the driver to confirm safety by instructing the alarm processing circuit 23 at the timing of rising of the vehicle speed or the like. Therefore, in this embodiment, it is possible to prevent an accident caused by ignoring the stop once before the level crossing 72.
[0111]
Furthermore, in this embodiment, since the traveling environment detection circuit 21 as the traveling environment detection means CL3 includes the NAVI, it is possible to reliably detect the level crossing 72 present in the traveling lane 71 from the host vehicle 41 to a predetermined distance ahead. It is also possible to detect the stop position at which the vehicle should stop at the level crossing 72.
[0112]
For this reason, in the present embodiment, it is possible to reliably detect whether or not the own vehicle 41 has stopped at the stop position, and whether or not the driver has confirmed left and right safety when stopping at the stop position. Can also be determined.
[0113]
FIG. 24 is a flowchart executed by the determination circuit when the automobile is moved backward in the present embodiment. However, the traveling environment detection circuit 21 shown in FIG. 2 includes a brake switch that detects whether or not the brake pedal is depressed, and a reverse switch that detects whether or not the gear position of the transmission is in a reverse position for reversing the vehicle. The brake switch detects whether or not the brake pedal is depressed, and the reverse switch detects whether or not the transmission gear position is the reverse position.
[0114]
When the vehicle is to be moved backward, as shown in FIG. 24, the determination circuit 22 of FIG. 2 inputs, as a reverse SW signal, whether or not the gear position of the transmission is the reverse position from the reverse switch in step S41. , Whether or not the brake pedal is depressed is input as a foot brake SW signal from the brake switch.
[0115]
In step S43, the determination circuit 22 determines whether or not the gear position of the transmission is in the reverse position based on the reverse SW signal input in step S41. If the gear position of the transmission is in the reverse position, the process proceeds to step S44. If the gear position of the transmission is not in the reverse position, the process returns to step S41.
[0116]
In step S44, it is determined whether or not the brake pedal is depressed based on the foot brake SW signal input in step S42. If it is detected that the brake pedal is not depressed and the foot brake is not activated, the process proceeds to step S45. If it is detected that the brake pedal is depressed and the foot brake is activated, the process returns to step S41.
[0117]
In step S45, it is determined whether or not the driver has confirmed safety based on the data from the facial feature quantity recognition circuit 15 shown in FIG. If the driver fails to confirm safety, the process proceeds to step S46, and if the driver confirms safety, the process returns to step S41.
[0118]
In step S46, the determination circuit 22 outputs a signal to the alarm processing circuit 23 shown in FIG. 2, and immediately generates an alarm such as a sound or light that prompts the driver to confirm safety from the alarm processing circuit 23.
[0119]
As described above, when the vehicle is moved backward in the present embodiment, the determination circuit 22 determines whether or not the transmission gear position is in the reverse position, and when the transmission gear position is in the reverse position, Determine whether the brake pedal is depressed, and if the brake pedal is released and no brake pedal depression is detected, and the foot brake is not activated, determine whether the driver has confirmed safety. If the driver fails to confirm the safety, the alarm processing circuit 23 is instructed to immediately generate an alarm prompting the driver to confirm the safety from the alarm processing circuit 23.
[0120]
For this reason, in this embodiment, if the driver fails to confirm safety when the vehicle is moving backward, an alarm can be issued immediately after the foot brake is released to prompt the driver to confirm safety. Accidents can be prevented without fail.
[0121]
When the vehicle is to be moved backward, the driving environment detection circuit 21 includes an accelerator switch for detecting whether or not the accelerator pedal is depressed, instead of the brake switch, and detects whether or not the accelerator pedal is depressed by the accelerator switch. You may make it do.
[0122]
In this case, the determination circuit 22 inputs the presence / absence of depression of the accelerator pedal from the accelerator switch as an accelerator SW signal in step S42 of FIG. 24, and in step S44, the accelerator pedal of the accelerator pedal is input based on the accelerator SW signal. The presence or absence of depression is determined, and if the accelerator pedal is depressed even a little, the process proceeds to step S45, and if the accelerator pedal is not depressed, the process returns to step S41.
[0123]
In this way, if the driver fails to confirm safety after the transmission gear position is in the reverse position and before the accelerator pedal is depressed, an alarm is issued immediately after the accelerator pedal is depressed. It is possible to prompt the driver to confirm the safety and to prevent accidents when the vehicle moves backward.
[0124]
As described above, in this embodiment, the image input unit CL1 inputs the face image of the driver of the automobile 41, and the feature amount detection unit CL2 performs image processing on the face image input by the image input unit CL1. The facial feature amount is detected, and the traveling environment detection means CL3 detects the traveling environment of the host vehicle.
[0125]
In the present embodiment, the determination circuit 22 that constitutes the travel environment determination unit CL4 and the safety confirmation determination unit CL5 requires the travel environment of the automobile 41 to confirm safety based on the detection data of the travel environment detection unit CL3. When it is determined whether or not the vehicle is in a driving environment that requires safety confirmation, it is determined whether or not the driver has confirmed safety based on the detection data of the feature amount detection means CL2, and the driver If it is determined that the safety check has been neglected, the safety promoting means CL6 is instructed to issue an alarm that prompts the driver to confirm the safety.
[0126]
Therefore, in the present embodiment, when the driver fails to confirm safety in a traveling environment where safety confirmation is required, the driver can be encouraged to maintain safety.
[0127]
In the present embodiment described above, an alarm processing circuit 23 including an alarm device that issues an alarm to the driver with sound, light, or the like is employed as the safety promoting means CL6. However, the safety promotion means CL6 is not limited to such an alarm processing circuit 23. For example, the safety promotion means CL6 brakes the host vehicle 41 with an automatic brake or the like based on the determination that the safety confirmation by the safety confirmation determination means CL5 is neglected. It may be one that promotes retention.
[0128]
If the vehicle 41 is urged to maintain safety by braking the own vehicle 41 with an automatic brake or the like as the safety promoting means CL6, the vehicle is decelerated when the course is changed while traveling, and once stopped at a railroad crossing or the like. By maintaining it, safety can be further improved. Of course, it is easy to combine an automatic brake or the like with an alarm.
[Brief description of the drawings]
FIG. 1 is a functional block diagram showing an embodiment of the present invention.
FIG. 2 is a configuration block diagram showing the configuration of an embodiment of the present invention.
FIG. 3 is a flowchart showing image processing of a driver's face image.
FIG. 4 is an explanatory diagram illustrating setting of a pixel column of a driver's face image.
FIG. 5 is a flowchart showing step S140 in FIG.
6 is a graph showing the density of the pixel row Xa in FIG. 4, where (a) shows the arithmetic mean density value, and (b) shows the differential value of the arithmetic mean density value.
7 is a graph showing the density of the pixel row Xb in FIG. 4, wherein (a) shows the arithmetic mean density value, and (b) shows the differential value of the arithmetic mean density value.
FIG. 8 is a graph showing the distribution of extracted points.
FIG. 9 is a flowchart showing step S150 in FIG.
FIG. 10 is a graph showing identification of candidate points from extracted points, where (a) shows grouping of extracted points and (b) shows identification of candidate points.
FIG. 11 is a flowchart showing step S160 in FIG. 3;
FIG. 12 is an explanatory diagram showing a face image of a driver's eyes and eyebrows.
13 is an explanatory diagram showing setting of a predetermined area shown in FIG. 12. FIG.
FIG. 14 is an explanatory diagram of a processed image showing a state in which the driver looks at the left door mirror.
FIG. 15 is an explanatory diagram of a processed image showing a state where the driver looks at the right door mirror.
FIG. 16 is an explanatory diagram of a processed image showing a state in which the driver looks obliquely backward.
FIG. 17 is a flowchart for determining whether or not a driver has confirmed safety.
FIG. 18 is an explanatory diagram showing a case where a vehicle changes lanes.
FIG. 19 is a flowchart executed by a determination circuit when a lane is changed.
FIG. 20 is an explanatory diagram showing a case where an automobile passes through an intersection.
FIG. 21 is a flowchart executed by a determination circuit when passing an intersection;
FIG. 22 is an explanatory diagram showing a case where an automobile passes through a railroad crossing.
FIG. 23 is a flowchart executed by a determination circuit when passing a railroad crossing.
FIG. 24 is a flowchart executed by the determination circuit when the vehicle is moving backward.
[Explanation of symbols]
CL1 Image input means
CL2 feature quantity detection means
CL3 Running environment detection means
CL4 traveling environment judgment means
CL5 Safety confirmation judgment means
CL6 Safety promotion means
11 TV camera
12 AD converter
13 Image memory
14 Image data arithmetic circuit
15 Feature recognition circuit
21 Driving environment detection circuit
22 Judgment circuit
23 Alarm processing circuit
33 eyes
34 windows (predetermined area)
41 Automobile (own vehicle)
63 Intersection
72 level crossing

Claims (3)

An image input means for inputting the face image of the driver of the car, a feature amount detecting means for detecting and storing a face feature value by performing image processing on the face image input by the image input means, and a traveling environment of the host vehicle A traveling environment detecting means for detecting, a traveling environment determining means for determining whether or not the traveling environment of the host vehicle is an environment that requires safety confirmation based on detection data of the traveling environment detecting means, and the traveling environment determining means; When it is determined that the driving environment requires safety confirmation, safety confirmation determination means for determining presence / absence of driver's safety confirmation based on detection data of the feature amount detection means, and safety by the safety confirmation determination means A safety promotion means for encouraging the driver to maintain safety based on the determination that the confirmation is neglected,
The traveling environment detection means detects a change in the steering angle of the host vehicle and a change in the course of the host vehicle by the operation of the direction indicator, and the traveling environment determination unit needs to confirm safety when the direction indicator is operated. When the driving environment is determined and the safety check determining means determines that the driver has failed to check the safety between the start of the operation of the direction indicator and the change of the steering angle, the safety check determining means prompts the driver to maintain safety. A safety promoting device for an automobile, characterized by instructing the safety promoting means. An image input means for inputting the face image of the driver of the automobile, a feature amount detecting means for detecting and storing a face feature amount by processing the face image input by the image input means, and a traveling environment of the host vehicle A traveling environment detecting means for detecting, a traveling environment determining means for determining whether the traveling environment of the host vehicle is an environment requiring safety confirmation based on detection data of the traveling environment detecting means, and the traveling environment determining means; When it is determined that the driving environment requires safety confirmation, safety confirmation determination means for determining the presence or absence of a driver's safety confirmation based on detection data of the feature amount detection means, and safety by the safety confirmation determination means A safety promotion means for encouraging the driver to maintain safety based on the determination that the confirmation is neglected,
The travel environment detection means detects whether or not the brake pedal is depressed and whether or not the gear position of the transmission is in the reverse position, and the travel environment determination means determines that the safety condition is required when the vehicle is in the reverse position. The safety confirmation determining means determines that if the driver has neglected the safety check from the time when the gear position of the transmission is in the reverse position until the absence of depression of the brake pedal is detected, A safety promoting device for an automobile, wherein the safety promoting means is instructed to promote safety maintenance. An image input means for inputting the face image of the driver of the automobile, a feature amount detecting means for detecting and storing a face feature amount by processing the face image input by the image input means, and a traveling environment of the host vehicle A traveling environment detecting means for detecting, a traveling environment determining means for determining whether the traveling environment of the host vehicle is an environment requiring safety confirmation based on detection data of the traveling environment detecting means, and the traveling environment determining means; When it is determined that the driving environment requires safety confirmation, safety confirmation determination means for determining the presence or absence of a driver's safety confirmation based on detection data of the feature amount detection means, and safety by the safety confirmation determination means A safety promotion means for encouraging the driver to maintain safety based on the determination that the confirmation is neglected,
The travel environment detection means detects whether or not the accelerator pedal is depressed and whether or not the transmission gear position is the reverse position, and the travel environment determination means determines that the safety condition is required when the vehicle is in the reverse position. The safety confirmation determining means determines if the driver has neglected the safety check from the time when the gear position of the transmission is in the reverse position until the presence of depression of the accelerator pedal is detected. A safety promoting device for an automobile, wherein the safety promoting means is instructed to promote safety maintenance.

Images