KR20070005553A - Imaging system - Google Patents

Abstract

An imaging system comprises an IR lamp irradiating infrared light, a CCD camera (5) for imaging a place irradiated by the IR lamp and converting the image into an electric signal, and an image processing unit (7) capable of continuously and periodically outputting images of different exposure amounts by varying the signal storage time of the CCD camera (5) at a specified period. The image processing unit (7) extracts a mass of high luminance having a spread of intermediate luminance on the periphery of the images outputted periodically on one hand, and controls the signal storage time of the images periodically outputted according to the extent of the mass of high luminance on the other hand. ® KIPO & WIPO 2007

Description

Imaging System

The present invention relates to an imaging system using a CCD camera or the like.

22 shows an imaging system according to the prior art. In FIG. 22, a charge-coupled device (CCD) 101 is provided as an imaging means, and a DSP (Digital Signal Processor) 103 and a CPU (central processing unit) 105 are provided as an image processing unit.

The CPU 105 and the DSP 103 are connected through a multiplexer 107. The signal from the shutter speed setting switch 109 is input to the CPU 105. The shutter speed setting switch 109 can set the shutter speed for the ODD field and the shutter speed for the EVEN field, respectively. The CPU 105 reads the shutter speed set by the shutter speed setting switch 109, encodes the shutter speed read for a given field, and outputs the encoded shutter speed. The DSP 103 outputs the field pulse signal shown in FIG. When the field pulse signal is high, the multiplexer 107 provides the shutter speed for the EVEN field to the shutter speed setting terminal of the DSP 103. When the field pulse signal is low, the multiplexer 107 provides the shutter speed for the ODD field to the shutter speed setting terminal of the DSP 103. In this way, the imaging system of Fig. 22 can set different shutter speeds for each field.

There is a CCD camera employing the same shutter speed for the ODD field and the EVEN field. Fig. 24 shows an example of an image taken by this type of CCD camera. The image contains a bright light source with a dark periphery. In this image, the bright light source and its borders become invisible due to halation.

The CCD camera used for image capturing in FIG. 24 is mounted on a vehicle equipped with an infrared (IR) lamp as infrared irradiation means. During night driving of the car, the car can irradiate infrared light from the IR lamp toward the front of the car, and can capture an image as shown in FIG. 24 with the CCD camera. The bright light source shown in FIG. 24 is the headlight of the car approaching on the opposite side, and the periphery of the bright light source is hidden by halation. If at night the bright light source is in the dark surroundings, the CCD camera employing the front photometry system is dominated by the dark surroundings, thus calculating the long exposure time, i.e. the slow shutter speed. This shutter speed is too slow for the bright light source, resulting in halation.

The shutter speed can be made faster to suppress the halation. However, this creates a problem that, as shown in Fig. 25, the dark part of the surrounding becomes darker and the background becomes invisible at all.

If there is a reflector such as the sign of Fig. 26, the shutter speed is controlled to be slow because of the above mark, so that the mark is not visible in the same way.

In order to overcome this problem, there is a double exposure control that changes the shutter speed for each field. This controls to output a bright image and a dark image alternately. Bright pictures (eg, EVEN fields) can illuminate dark areas, and dark pictures (eg, ODD fields) can illuminate bright areas that can cause halation in bright pictures.

Bright and dark images can be output alternately and displayed on the monitor as clear images.

However, there is a problem that the double exposure control causes flicker on the monitor by controlling to output a bright image and a dark image (field) alternately.

Moreover, there is an imaging device shown in Fig. 27 described in JP-A-7-97841. This imaging apparatus is equipped with the camera part 113 provided with the imaging element 111, and the processing part 115. As shown in FIG.

FIG. 28 shows a conceptual diagram of image processing by the image capturing apparatus of FIG. 27. In FIG. 28, a through image is an image which is directly output from the image pickup device 111 of the camera unit 113, and a memory image is a field immediately before being stored in the image memory 117 once. Says burns.

In FIG. 28, each ODD field is set to a fast shutter speed, and each EVEN field is set to a slow shutter speed. In each ODD field, the through image represents a crushed black main subject, and in each EVEN field, the through image represents a saturated white background. Since each memory picture is located one field later, the crushed black or saturated white color is opposite to the corresponding through picture. By properly combining the through image and the memory image, an appropriate output image shown at the bottom of Fig. 28 can be obtained.

However, since the synthesis of the through image and the memory image partially extracts and superimposes the through image and the memory image, images of different exposure amounts are mainly combined together. Therefore, although flickering such as the simple double exposure control can be reduced, there is a problem that the boundary between the synthesized through image and the memory image becomes unnatural.

An object of the present invention is to provide an image capturing system that enables a clearer image output.

In order to achieve the above object, a first aspect of the present invention is to provide an infrared light irradiation means for irradiating infrared light, a place irradiated by the infrared light irradiation means, and to pick up the captured image by an electric signal. And an image processing unit configured to periodically change the signal accumulation time of the image pickup means and to periodically output images having different exposure amounts. The image processing unit extracts a high-brightness block surrounded by a medium-brightness area from a first image of the periodically output image, and the degree of the medium luminance area is extracted. According to the degree, the signal accumulation time of the second image imaged is controlled.

According to the first aspect, the infrared light irradiation means irradiates infrared light, the image pickup means images the place irradiated by the infrared light irradiation means and converts it into an electrical signal, and the image processing unit is a signal of the imaging means. By accumulating the accumulation time periodically, images of different exposure amounts are periodically outputted.

According to the first aspect, the image processing unit extracts a high luminance block surrounded by a medium luminance region from the first image of the periodically output image, and according to the degree of the medium luminance region, Control signal accumulation time.

By this control, even if strong light from the headlights of the opposite deviation enters the imaging means, for example, the invention of the first aspect is gradually darkened around the high luminance block by the strong light in the captured image. By removing or suppressing the lost area, even if there is a pedestrian or obstacle near the strong light, the image of the pedestrian or obstacle can be clearly captured.

According to a second aspect of the present invention, the image processing unit divides the first image into a high luminance block, a medium luminance block, and a low luminance block, and according to the number of medium luminance blocks around a group of high luminance blocks, the second image. Image signal accumulation time of an image is controlled.

Therefore, according to the number of heavy luminance blocks around a group of high luminance blocks, the invention of the second aspect can reliably grasp the degree of the heavy luminance block and control the signal accumulation time of the second image.

According to a third aspect of the present invention, the image processing unit divides the first image into a plurality of blocks to obtain a luminance average value of each block, and according to the luminance average value of the blocks and two thresholds. The blocks are classified into high luminance blocks, medium luminance blocks, and low luminance blocks.

Therefore, the invention of the third aspect can improve the image processing time as compared to the technique of processing image pixels by pixels.

According to a fourth aspect of the present invention, the image processing unit divides the first image into a plurality of blocks, and divides pixels in each block into high luminance pixels, medium luminance pixels, and low luminance pixels by two threshold values. Classify and find the maximum number of the high luminance pixels, the medium luminance pixels, and the low luminance pixels in each block, determine the luminance level of the pixel having the maximum number as the luminance level of the block, and determine the luminance of the determined block. According to the level, the blocks are classified into high luminance blocks, medium luminance blocks, and low luminance blocks.

The invention of the fourth aspect can enhance the accuracy by processing the image pixels by the pixels.

According to a fifth aspect of the present invention, the image processing unit finds the number of the medium luminance blocks around each of the high luminance blocks, finds the maximum number of the number of the medium luminance blocks around the high luminance block, and according to the maximum number. The image signal accumulation time of the second image is controlled.

According to the fifth aspect of the invention, the halation can be easily specified, and rapid image processing can be performed.

According to a sixth aspect of the present invention, the image processing unit includes a number of high luminance blocks forming a group, a number of heavy luminance blocks around the group, and a reference number of heavy luminance blocks associated with the group. The image signal accumulation time of the second image is controlled according to these numbers.

According to the invention of the sixth aspect, the halation can be appropriately specified, and more accurate image processing can be performed.

According to a seventh aspect of the present invention, the image processing unit identifies a high luminance block and finds a periphery of the high luminance block with respect to the medium luminance block and the high luminance block, and the high luminance blocks are found. Group into blocks

According to the seventh aspect of the invention, the high luminance block can be extracted and controlled accurately and quickly.

According to the eighth aspect of the present invention, the infrared light irradiation means, the imaging means, and the image processing unit are provided in an automobile. The infrared light irradiation means irradiates infrared light to the outside of the vehicle. The imaging means picks up the outside of the vehicle.

Therefore, even if there is halation caused by headlights of opposite cars, for example, the invention of the eighth aspect can remove or suppress the gradually darkening area around the halation. Therefore, even if there is a pedestrian or obstacle near the halation, the image of the pedestrian or obstacle can be clearly captured.

1 is a conceptual diagram of a vehicle provided with an imaging system according to an embodiment of the present invention.

2 is a block diagram showing a camera and an image processing unit of the imaging system.

3 is a flowchart showing exposure switching control according to an embodiment of the present invention.

4 is an image diagram showing a light source in an image picked up under simple control.

5 is a graph showing a change in luminance along a horizontal line across the light source of FIG. 4.

6 is an image diagram showing a reflector picked up under simple control.

FIG. 7 is a graph illustrating a change in luminance along a horizontal line crossing the reflector of FIG. 6.

8 is a schematic diagram of dividing luminance data of an EVEN field into several blocks according to an embodiment of the present invention.

Fig. 9 relates to an embodiment of the present invention and is a diagram showing block color classification by gray ratios.

10 is a conceptual diagram illustrating block color classification according to an embodiment of the present invention.

Fig. 11 is a conceptual diagram showing an intrablock search procedure in accordance with an embodiment of the present invention.

Fig. 12 is a diagram showing an output image including a bright light source whose outer periphery is found.

FIG. 13 relates to an embodiment of the present invention, and is an image diagram showing a result of inspection of the image of FIG. 12 in which the outer periphery of the bright light source is displayed in three colors.

Fig. 14 relates to an embodiment of the present invention, which shows the relationship between the number of standard blocks and the number of white blocks, in which Fig. 14 (a) shows one white block and Fig. 14 (b) shows two white blocks. 14 (c) is a conceptual diagram including three white blocks.

15 is a conceptual diagram illustrating the number of halation detection blocks in accordance with an embodiment of the present invention.

FIG. 16 is a diagram illustrating an output image showing a reflection object that can reflect light and a light source.

FIG. 17 relates to one embodiment of the present invention and is a processed image diagram showing the results of color classification performed on the image of FIG. 16.

18 is a diagram showing an exposure difference between an EVEN field and an ODD field according to an embodiment of the present invention.

Fig. 19 relates to an embodiment of the present invention and is a state transition diagram of halation intensity.

Fig. 20 relates to one embodiment of the present invention and is a processed image diagram of an example in which an object is seen in halation.

Fig. 21 relates to one embodiment of the present invention and is a processed image diagram showing an example of the surroundings irrespective of the reflecting material.

Fig. 22 is a block diagram showing an imaging system according to a conventional example.

Fig. 23 relates to a conventional example and is an output diagram of a field pulse.

24 relates to a conventional example, and is an output image diagram showing an example in which the light source and its surroundings are not visible due to halation.

Fig. 25 relates to a conventional example and is an output image diagram including portions that are not visible by halation.

Fig. 26 relates to a conventional example and is an output image diagram showing an example in which the reflector and its surroundings are not visible.

27 is a block diagram showing an imaging system according to another conventional example.

FIG. 28 is a view showing an image provided by the conventional example according to FIG. 27.

An imaging system according to an embodiment of the present invention will be described. 1 is a conceptual diagram of a vehicle to which an imaging system according to an embodiment of the present invention is applied, FIG. 2 is a block diagram of the imaging system, and FIG. 3 is a flowchart showing exposure switching control according to the embodiment.

In Fig. 1, a motor vehicle 1 includes an IR lamp 3 provided as infrared irradiation means for irradiating infrared rays, a CCD camera 5 as an image pickup means, an image processing unit 7, and a head-up. up) a display 9.

The IR lamp 3 irradiates infrared light toward the front of the vehicle 1 in the traveling direction for imaging at night or when the circumference of the vehicle 1 is dark. The CCD camera 5 photographs the front of the vehicle 1 as a photosensitive element with infrared light emitted from the IR lamp 3, and converts the captured image into an electrical signal. More specifically, a photodiode in the CCD camera 5 converts the image into an electrical signal. The image processing unit 7 changes the signal accumulation time of the CCD camera 5 at a predetermined cycle, and periodically outputs images with different exposure amounts continuously.

The signal accumulation time is set for all the pixels. Changing the signal accumulation time by a predetermined period means changing the number of pulses for releasing unnecessary charge accumulated in each pixel. This operation is an electronic shutter operation. In addition, periodically outputting images with different exposure amounts continuously sets the shutter speed for each ODD field and EVEN field by the electronic shutter operation, and displays images of each field read at each shutter speed, for example, 1/60 second. It means to output successively at intervals.

In high-speed shutters with faster shutter speeds, darker parts are less visible, while brighter parts are more clearly visible, while slower shutter speeds are slower, brighter parts are saturated and fly away, and darker parts are more clearly visible.

The image processing unit 7 extracts, from the first image, a high luminance block having a medium luminance at the outer circumference, and controls the signal accumulation time of the second image according to the degree of the medium luminance.

In Fig. 2, the CCD camera 5 and the image processing unit 7 include a CCD 5a, an analog front end (AFE) 11, a DSP 13, a RAM 15, a CPU 17, and the like. Doing.

The CCD camera 5 includes a part of the CCD 5a, the AFE 11, the DSP 13, and the CPU 17. As shown in FIG. The image processing unit 7 includes a part of the DSP 13, a RAM 15, and a CPU 17.

The AFE 11 amplifies the output signal of the CCD 5a and converts the amplified signal into a digital signal.

The DSP 13 is a digital signal processor, which generates timing signals for the CCD 5a and AFE 11 operation, and converts the signals from the CCD 5a through the signal conversion, video signal generation, and the AFE 11. Γ conversion of signals, enhancement processing, digital signal amplification processing, and the like are performed.

The RAM 15 is a memory and temporarily stores luminance (= density) data of an EVEN field image output from the DSP 13.

The CPU 17 performs various operations and controls the shutter speed for each ODD field and EVEN field by the same configuration as described in FIG. 22. That is, for the EVEN field, the CPU 17 calculates an optimal exposure condition according to the average brightness of the EVEN field, and amplifies control for the AFE 11 and electrons for the CCD 5a according to the calculated conditions. Shutter operation control is performed.

The operation of the imaging system of FIG. 2 will be described.

First, the CPU 17 sets an initial shutter speed and outputs a shutter speed control signal on the ODD field side and a shutter speed control signal on the EVEN field to the DSP 13.

In the DSP 13, timing signals for the operation of the CCD 5a and the AFE 11 are generated. Based on this timing signal, imaging by the CCD 5a is performed, and the photodiodes of all the pixels of the CCD 5a accumulate signal charges. On the ODD field side alternately arranged in the middle of the EVEN field in the vertical direction, the signal charges of every odd-numbered photodiode (pixel) in the vertical direction are read at the set shutter speed. On the EVEN field side, the signal charge of every even photodiode (pixel) in the vertical direction is read at the set shutter speed.

The signal charge read from the CCD 5a is amplified by the AFE 11 and converted into a digital signal and provided to the DSP 13. In the DSP 13, signal conversion, video signal generation, gamma conversion, enhancer processing, digital signal amplification processing, and the like of the provided signal are performed.

The luminance data of the EVEN field image output from the DSP 13 is temporarily stored in the RAM 15.

The CPU 17 calculates an optimal exposure condition for the EVEN field from the average concentration, and controls the electronic shutter for the CCD 5a through the DSP 13 according to the calculated condition.

The CPU 17 calculates exposure conditions for the ODD field by exposure switching control shown in the flowchart of FIG. 3.

In step S1 of exposure switching control in Fig. 3, luminance data in units of EVEN field blocks is taken out. More specifically, step S1 divides the luminance data of the EVEN field stored in the RAM 15 into several blocks and calculates an average luminance of each block.

In step S2, a process of "block-level digitized data conversion" is executed. This step uses two thresholds to convert the average luminance of each block provided in step S1 into digitized data.

In step S3, the process of "high brightness block detection" is performed. In this step, the quantization data of each block is examined, and the mass of a high brightness block is detected.

In step S4, the process of "high brightness block grouping" is performed. In this step, adjacent high luminance blocks are combined (grouped), and the number of blocks in each group is detected as the size of each group.

In step S5, the process of "medium intensity block detection" is performed. In this step, a group of heavy luminance blocks around the high luminance block group is detected.

In step S6, the processing of "halation level calculation" is performed based only on the size of the high luminance group, the size of the medium luminance group, or the size of the medium luminance group. In this step, the maximum halation level (intensity) is detected in the EVEN field.

In step S7, the exposure target value for the ODD field is determined. In this step, the exposure amount of the ODD field is calculated according to the halation level of the EVEN field. After this, the process proceeds to the next EVEN field.

According to the exposure conditions calculated through the steps of FIG. 3, the electronic shutter 1 of the CCD 5a, the AGC gain of the AFE 11, the digital gain of the DSP 13, and the like are controlled and obtained. To optimize the brightness.

On the other hand, the quantization data calculation in step S2 may be based on the attribute of the pixel which occupies the most part in the predetermined block instead of the average luminance of the block.

When the CCD camera 5 of FIG. 1 receives a strong light, for example, the headlights of the opposite car, the above-described control does not affect the strong light without degrading the brightness of the image of the dark part around the headlights. It can be suppressed.

The CCD camera of an imaging system used for automobiles generally employs an interlace method as an image method. The interlacing method consists of two fields, an EVEN field and an ODD field, and the two fields are alternately outputted in time, so that the image of the viewer has a constant resolution.

The conventional CCD camera calculates exposure conditions from the average luminance of the EVEN field or the ODD field. The exposure condition can output an image of optimal brightness on the monitor by determining the electronic shutter speed of the CCD, the amplification factor (AGC gain) in the AFE, and the digital amplification factor of the DSP.

In a typical CCD camera, the calculated exposure condition is applied to any of the EVEN field / ODD field, and as a result, both fields output images having the same brightness. When there is a strong light source such as a headlight of the opposite vehicle approaching at night, in the conventional technique, the exposure conditions are determined according to the overall luminance average value. As a result, in the above conventional technique, a strong light source and its surroundings are often output as a white saturated image (so-called halation).

An example of halation is shown in FIG. 4. In this figure, the strong light source and its surroundings are saturated white, and the saturated portions thereof are scattered radially. FIG. 5 is a graph showing luminance levels along a line across the strong light source of FIG. 4. In Fig. 5, the strong light source and its surroundings are saturated to the maximum brightness, and the brightness gradually decreases away from the strong light source.

If the pedestrian is in the saturation region, it is impossible for the CCD camera to grasp the image of the pedestrian. Although the center of the strong light source is saturated white, the surroundings or between the headlights should not be saturated so that the image of the pedestrian can be captured.

6 shows an example in which the reflector is imaged. In FIG. 6, the reflector is a billboard, a road sign, a signboard, or the like, which reflects light from the headlight. FIG. 7 is a graph illustrating luminance levels along a line crossing the reflector of FIG. 6. In Figure 7, the reflector is white saturated. However, there is no halation around the reflector. That is, the luminance curve of FIG. 7 drops sharply from the peak of the reflector. If there is a pedestrian near the reflection, the pedestrian can be clearly identified as an image. In this case, it is not necessary to take measures against halation. That is, it is not necessary to suppress the exposure amount of the ODD field than the exposure amount of the EVEN field. On the contrary, considering that the amount of light of the subject is small at night, as in the EVEN field, it is preferable that the exposure amount be sufficiently secured to easily recognize an object such as a pedestrian.

According to the present invention, each EVEN field is processed according to the flowchart of FIG. 3 to detect halation and to detect an optimal exposure condition according to the technical concept of the present invention, so that a dark object at night is brighter. For each ODD field, an exposure difference is set on the basis of the luminance data obtained in the previous EVEN field, so as to display an image which minimizes the influence of strong light.

By combining the images of the ODD field and the EVEN field with these other characteristics, the present invention clearly does not cause halation even when receiving a strong light at night, and keeps the bright part of the light bright so that each object can be clearly seen. The image shown can be output.

Here, a series of processes until detecting the halation level from the luminance data of the EVEN field, calculating the exposure condition according to the halation level, and outputting the calculation result will be described in detail.

The block classification of the EVEN field will be described. This process is executed in step S1 of FIG. The luminance data of the EVEN field is taken out of the DSP 13 and stored in the RAM 15. The luminance data is composed of, for example, 512 dots x 240 lines. This data is divided into, for example, 64 x 60 blocks, each block having 8 dots x 4 lines, as shown in FIG.

Then, the average value of the luminance data of each block is calculated. That is, the average of pixel values of all pixels (8x4 pixels) of each block is calculated. This calculation is performed in step S1 of FIG.

Thereafter, the luminance average of each block is converted into digitized data using two thresholds. This conversion is performed in step S2 of FIG. Each luminance value is represented by 8 bits, for example. In this case, the lowest luminance is 0 and the highest luminance is 255. For example, the white threshold is set to 220 (or higher) and the black threshold is 150 (or lower), and the middle is set as gray. Each block is classified as follows according to its average brightness:

White when mean luminance ≧ white threshold; or

Gray when white threshold> average luminance≥black threshold; or

Black when average luminance <black threshold.

Instead of converting the average luminance of each block to ternary data using two thresholds, each pixel in each block is converted to ternary data using the same threshold, resulting in white (high brightness) pixels, gray (medium brightness) The number of pixels, black (low luminance) pixels can be calculated, and the color of the maximum number of pixels can be specified for the block.

9 is an example showing classification based on pixels provided with digitized data. Each pixel in a given block is classified into white, gray and black according to the number of gray pixels. When the ratio of gray pixels of the block is 50% or more, the block is classified as gray. When the ratio of gray pixels is less than 50%, it is classified as white or black. In FIG. 10, an example of the block which has 32 (8x4) pixel is shown. Since the gray fraction of this block is at least 50%, the block is classified as gray.

Instead of dividing the field into blocks, each pixel can be studied for halation detection and exposure calculation.

After each block is converted into digitized data, group processing is performed in step S4 of FIG. The group processing detects mass of white (high brightness) blocks.

In Fig. 8, the block checks from the first block (0,0) to the last block (63,59) to find the white blocks sequentially from the first line. In each line, the check is performed from the left block of FIG. 8 toward the right block.

If one white block is found, then there are white blocks for the surrounding eight blocks of the found white block as shown in FIG. In FIG. 11, the check is performed from the left block “1” toward the lower left block “8” in the clockwise direction. The white blocks found are connected in series to form a group of white blocks.

FIG. 12 is an output image diagram including a strong light source irradiated around its circumference, and FIG. 13 shows an image processed through steps S1 to S4 described above. The strong light source of Figure 12 comes from the headlights. In FIG. 13, the strong light source of FIG. 12 is surrounded by continuous gray blocks. Inside the gray block boundary, there are only white blocks, which form one group.

Then, the halation is detected in step S5 of FIG. Halation contains saturated centers due to strong light, and it gradually darkens around it. The center of the halation is a group of white blocks, around which gray blocks are formed.

To detect halation, gray blocks around a group of white blocks are detected and the number of gray blocks is counted.

14 shows an ideal or typical example of halation with a group of white blocks with gray blocks around them. In the example of the halation of FIG. 14 (a), one white block group is composed of one white block and eight gray blocks around it. In the example of the halation of Fig. 14 (b), one white block group is composed of two white blocks and ten gray blocks around it. In the example of the halation of Fig. 14 (c), one white block group is composed of three white blocks and twelve gray blocks around it. This gray block number is provided as a reference block number when calculating the halation level according to the second calculation method described later.

The halation level is calculated based on the group of white blocks and the gray blocks around them in step S6 of FIG.

The first method of calculating the halation level is to find the maximum number of gray blocks around one white block group among the white block groups, and determine the maximum number as the halation level.

The second method of calculating the halation level is a method of examining the size of a predetermined white block group and the halation probability of the group.

The first method will be described.

The first method counts the number of gray blocks around each white block group (source sample). From the number of grayed out blocks, the maximum number is selected as the halation level.

Halation Level = 1 Maximum number of gray blocks around a white block group in a picture

15 shows an example of a white block group having one white block. In this example, the number of gray blocks around the white block group is seven, so the halation level is seven.

16 shows an initial image that includes a reflector and a light source having halation. FIG. 17 shows a processing image in which blocks are formed from ternary data created from the image of FIG. The image of FIG. 17 includes many white blocks and white block groups. Gig whiteblocks and white block groups are examined to determine the number of gray blocks nearby, and the maximum number of gray blocks is selected as the halation level.

According to the first method, the image of FIG. 17 is analyzed as follows:

0 gray blocks around the large signboard in the center above the image

0 gray blocks around the small signboard in the left part of the image

2 gray blocks around the front vehicle taillight in the center of the image

4 gray blocks around a street lamp on the top right of the image

32 gray blocks around the car headlights across the bottom right of the image

As a result, 32 is the maximum number of gray blocks of the picture of Fig. 17, and therefore its value is selected as the halation level.

A second method of calculating the halation level based on the size and probability of the white block group is described.

The number of gray blocks around a given group of white blocks is actually counted. The number of white blocks in the white block group is counted and, according to this number, the standard number of gray blocks (FIG. 14) is calculated. According to the actually calculated number of gray blocks and the standard number of gray blocks, the halation level of the white block group is calculated as follows:

Halation Probability (%)

         = (Actual number of gray blocks / standard number of gray blocks) × 100

The halation probability is multiplied by the number of white blocks of the white block group to provide the halation level of the white block group.

The image of FIG. 17 is analyzed as follows:

Large signage halation level (0/26) × 100 × 21 = 0

Small signage halation level (0/26) × 100 × 7 = 0

Halation level (2/8) × 100 × 1 = 25 of the taillight

Halation Level (4/18) × 100 × 8 = 178

Headlight halation level (32/37) × 100 × 43 = 3718

The maximum value of the halation levels is selected as the halation level of the image of FIG.

That is, the 3718 halation level around the headlights of the opposite vehicle is selected as the halation level of the image of FIG.

Thereafter, the exposure conditions for the ODD field are determined in step S7 of FIG.

According to the halation level of the EVEN field determined as described above, the exposure difference between the EVEN field and the ODD field is obtained, for example, from the table shown in FIG. In accordance with the exposure difference, an exposure condition for the ODD field is determined to suppress the halation.

The exposure difference is 0 dB when the halation level obtained according to the first method is in the range of STEP 0 (0 to 5) in FIG. 18, and the exposure difference is-when the halation level is in the range of STEP 6 (31 to 35). 12 dB. The exposure difference is 0 dB when the halation level obtained according to the second method is in the range of STEP 0 (0 to 500), and the exposure difference is -12 dB when the halation level is in the range of STEP 6 (3001 to 3500). .

If the halation level is in the range of STEP 0, there is no exposure difference between the EVEN field and the ODD field. If the halation level is in the range of STEP 6, the exposure amount of the ODD field is set to be about 12 dB smaller than the exposure amount of the EVEN field.

In this way, the halation level of a given EVEN field is classified into one of STEP O to STEP 10, and the exposure amount of the corresponding ODD field is reduced as the exposure difference indicated corresponding to the rightmost column of FIG.

As described above, the present invention performs double exposure control according to the level of halation, so that even if there is a strong light source such as a headlight of a car approaching under a dark environment such as at night, no halation occurs, Bright, excessively bright parts can be made darker so that they can be seen.

In practice, it is necessary to minimize the discomfort caused to the motor vehicle driver due to the change in luminance between images. For this purpose, as shown in FIG. 19, when strong light is received, the double exposure control accordingly is quickly performed, and when the light is weakened, the ODD field is gradually made brighter.

For example, when the automobile 1 (FIG. 1) turns around a corner and encounters the strong light of the headlight of the vehicle opposite, it makes the double exposure control according to the halation level immediately. In this case, when the opposite difference passes, the halation level falls to the range of STEP 0. If the double exposure control immediately follows the exposure range of STEP 0, there is a fear that the halation level of the image provided to the driver of the vehicle 1 changes abruptly, causing discomfort to the driver.

In order to avoid this, when the light incident on the CCD camera 5 (Fig. 1) is weakened, the image on the ODD field side is gradually brightened to minimize, eliminate or suppress discomfort.

An example is explained. When the vehicle 1 (Fig. 1) turns a corner, there exists a vehicle opposite, and suddenly the halt intensity of the headlight of the vehicle falls within the range of STEP 6 in Fig. 18. When the halation level is continued for at least two frames of the EVEN field, the exposure amount on the ODD field side is immediately controlled in accordance with the exposure difference (-12 dB) set for the halation level in STEP 6. The halation is weakened when the opposite car passes. When the halation level below STEP 6 has been continued for at least three frames of the EVEN field, the exposure amount for the ODD field is controlled according to the exposure difference (-10 dB) set for the halation range of STEP 5. Then, when the halation level below STEP 5 continues for at least three frames of the EVEN field, the exposure amount for the ODD field is controlled according to the exposure difference (-8 dB) set for the halation range of STEP 4. In this way, the exposure amount for the ODD field is gradually changed to the halation range of STEP 0 to gradually lighten the image on the ODD field side. Therefore, the driver of the vehicle 1 may not feel discomfort.

In this way, the imaging system according to one embodiment of the present invention can be controlled to switch the exposure according to the direct light and the reflected light. Even when the automobile 1 is directly subjected to strong light such as a headlight, the imaging system can remove or suppress the halo of strong light including a white saturation region of the center portion and a dimly dark surrounding. Therefore, even if there is a pedestrian or obstacle around the halation, the imaging system according to the present invention can clearly grasp the pedestrian or obstacle as shown in FIG.

When the headlights of the automobile 1 (FIG. 1) are illuminated, for example, with a signage, the reflected light reflected from the signage can reach the car 1, as shown in FIG. 21. In this case, only the reflector (signboard) itself becomes a white saturated image, but hardly halation occurs around it. In other words, the luminance curve of the reflector drops sharply from each edge of the peak (Fig. 7). Thus, the pedestrian or obstacle in the vicinity of the reflector can be clearly identified as an image. In this case, it is not necessary to change the exposure amount between the EVEN field and the ODD field. On the contrary, in order to clearly grasp the subject at night, it is necessary to secure a sufficient exposure amount for the ODD field and the EVEN field.

As described above, according to the imaging system of the embodiment of the present invention, it is possible to reduce halation caused by strong sources such as headlights of opposite cars, and to clearly project obstacles and pedestrians that may be around the halation. . With respect to reflections from billboards, signboards, road signs, etc., the above embodiments can maintain a sufficient exposure to provide bright images.

The image processing unit 7 according to the present embodiment divides an image of the EVEN field into a high luminance white block, a medium luminance gray block, and a low luminance black block, and divides the gray block around a group of white blocks of the EVEN field. The exposure amount of the ODD field is controlled according to the number.

Based on the number of gray blocks around the white block group of the EVEN field, the above embodiment can easily estimate the halation level of the white block group, and according to the halation level, it is appropriate to suit the exposure amount of the periodically output ODD field. Control.

The image processing unit 7 divides the EVEN field of the image into a plurality of blocks to calculate the luminance average value of each block, and classifies the blocks into white, gray, and black blocks using two luminance thresholds.

This technique allows for faster processing than exposing the luminance per pixel to divide the EVEN field into blocks.

The image processing unit 7 divides the EVEN field into a plurality of blocks, classifies the pixels of each block into white, gray, and black pixels using two thresholds, and counts the number of white, gray, and black pixels. Select the maximum number to be the color of the block.

This technique processes pixels on a pixel-by-pixel basis, so that accurate results can be obtained.

The image processing unit 7 can control the image signal accumulation time of the ODD field according to the maximum number of gray blocks around the white block group of the EVEN field.

Therefore, the halation can be easily specified, and the halation can be coped with quickly.

The image processing unit 7 each ODD field according to the number of white blocks in the white block group, the number of gray blocks around the interest white block group, and the standard block number of gray blocks corresponding to the white block group of the EVEN field. The image signal accumulation time can be controlled.

Therefore, the intensity of halation can be specified accurately, and the halation can be appropriately coped with.

The image processing unit 7 specifies a white block in the EVEN field of the image, and then finds a gray block around the white block. If another white block is found, the image processing unit 7 merges it with the preceding white block.

Therefore, the group of white blocks can be extracted accurately and quickly to control the halation associated with the white block group.

The imaging system according to the present embodiment includes the IR lamp 3, the CCD camera 5, and the image processing unit 7 mounted on an automobile. The IR lamp 3 irradiates infrared light to the front of the vehicle, and the CCD camera 5 can capture the front of the vehicle.

The imaging system according to the present embodiment can reduce halation caused by a strong light source such as headlights of opposite cars by removing or suppressing the area around the headlights in which the high brightness gradually changes to the low brightness. Therefore, even if there is a pedestrian or obstacle in the vicinity of the halation, the imaging system can clearly grasp its image.

The relationship between the EVEN field and the ODD field may be set in reverse. In other words, it is possible to obtain a halation intensity of the ODD field, to obtain an exposure difference between the ODD field and the EVEN field according to the halation intensity, and to suppress exposure in the EVEN field.

The DSP 13 may read charges of the ODD field and the EVEN field for each pixel or for each group of pixels.

In the above embodiment, the output image is displayed on the head-up display 9. Instead, the image may be configured to be displayed on a monitor provided in a car interior or the like. In the present embodiment, the IR lamp 3 is used to irradiate the front of the vehicle in the driving direction, but instead the rear or the side is irradiated, and the CCD camera 5 captures the rear or the side of the vehicle. You may.

The imaging system of the above embodiment can be applied to two-wheeled vehicles, ships, and the like, without being limited to automobiles. The imaging system may be configured as a stand-alone imaging system.

As described above, the imaging system according to the present invention irradiates infrared light to the front of the vehicle during the night driving of the vehicle, captures the front of the vehicle with a CCD camera or the like installed in the vehicle, and displays the state of the front of the vehicle as an image. I can figure it out.

Claims (8)

Infrared light irradiation means for irradiating infrared light, Imaging means for imaging a place irradiated by the infrared light irradiation means and converting the captured image into an electrical signal; And an image processing unit configured to periodically change a signal accumulation time of the imaging means periodically and periodically output images having different exposure values, The image processing unit extracts a high-brightness block surrounded by a medium-brightness area from a first image of the periodically output image, and the degree of the medium luminance area is extracted. and the signal accumulation time of the second image to be imaged is controlled in accordance with the degree. The method of claim 1, The image processing unit divides the first image into a high luminance block, a medium luminance block, and a low luminance block, and controls the image signal accumulation time of the second image according to the number of the medium luminance blocks around the group of the high luminance blocks. An imaging system, characterized in that. The method of claim 2, The image processing unit divides the first image into a plurality of blocks, obtains an average luminance value of each block, and divides the block into a high luminance block, a medium luminance block, and a block according to the luminance average value of the blocks and two thresholds. And a low luminance block. The method of claim 2, The image processing unit divides the first image into a plurality of blocks, classifies pixels in each block into high luminance pixels, medium luminance pixels, and low luminance pixels by two thresholds, and then stores the high luminance pixels in each block. Finding a maximum number of the medium and low luminance pixels and determining a luminance level of the pixel having the maximum number as the luminance level of the block, and determining the blocks according to the luminance level of the determined block. An imaging system characterized by being classified into a medium luminance block and a low luminance block. The method according to any one of claims 2 to 4, The image processing unit finds the number of the medium luminance blocks around each of the high luminance blocks, finds the maximum number of the neighboring medium luminance blocks, and sets the image signal accumulation time of the second image according to the maximum number. An imaging system, characterized in that for controlling. The method according to any one of claims 2 to 4, The image processing unit finds the number of high luminance blocks forming a group, the number of heavy luminance blocks around the group, and a reference number of the heavy luminance blocks associated with the group, according to the numbers. An image pickup system, characterized by controlling the image signal accumulation time of two images. The method according to claim 5 or 6, The image processing unit identifies the high brightness block, finds the outer periphery of the high brightness block with respect to the medium brightness block, and the high brightness block, and groups the found high brightness blocks into high brightness blocks. system. 8. The motor vehicle according to any one of claims 1 to 7, wherein the infrared light irradiation means, the image pickup means, and the image processing unit are provided in an automobile, The infrared light irradiation means irradiates infrared light to the outside of the vehicle, And the imaging means picks up the outside of the vehicle.

Images