JP5403779B2 - Lighting system - Google Patents

Description

  The present invention relates to a lighting system that controls blinking and dimming of a lighting fixture in accordance with the presence or absence of a person in a lighting space.

  2. Description of the Related Art Conventionally, there is known an illumination system that detects the presence or absence of a person in a room, the position of a person, etc. by an image sensor installed on the ceiling of a living room and controls blinking and dimming of a lighting fixture according to the detection result. .

  For example, as shown in FIG. 10, human detection by an image sensor compares two or more images captured in a room at a predetermined cycle, and can be regarded as a moving part, i.e., a fixed part. This is done by specifying the lump 101 of pixels.

  In addition, for two or more images that are periodically picked up, the first image is picked up with strong illumination, then the second image is picked up with weak light, and a difference area between the two images is extracted. Thus, a method for determining the presence or absence of a person has been proposed (see, for example, Patent Document 1).

  In these methods, since the presence or absence of a person cannot be determined if the person stops after moving, if the block of pixels that have moved previously stops or the difference area that can be extracted continuously is When it becomes impossible to extract, it is necessary to provide a determination circuit for determining that this is a person.

  However, in an indoor space such as a living room or a dining room, it is not always the person who moves the imaging range of the image sensor. For example, if there is a small animal such as a moving pet, a fan that swings, a device that moves and rotates, a curtain that shakes by the wind, etc., these are mistakenly determined to be humans, and as a result, the system controls appropriately There is a problem that it is not.

JP 2001-333420 A

  The present invention has been made in view of the above circumstances, and provides an illumination system that can accurately determine the presence or absence of a person in an illumination space and can appropriately control blinking and dimming of a lighting fixture. The purpose is to do.

The present invention is an illumination system that controls blinking and dimming of a lighting fixture according to the presence or absence of a person in the illumination space, an imaging device that sequentially images the illumination space in time series and outputs image data; A determination device that determines the presence or absence of a person in the illumination space based on the image data, and a lighting control device that controls blinking and dimming of the lighting fixture according to a determination result of the determination device, The determination device sequentially sets a difference between the time-series image data, a setting unit that sets first texture data to be canceled from the image data, a storage unit that stores the image data and the first texture data. By calculating, a calculation unit that extracts second texture data that moves in the illumination space, the first texture data, and the second texture data A determination unit that determines the presence or absence of a person in the illumination space in accordance with a match or a mismatch, and a display unit that displays an image captured by the imaging device, and the setting unit displays on the display unit The first texture data is set based on an image area that is specified for the image in the illumination space and includes a specific object that moves in the illumination space other than a person .

With this configuration, the first texture data is set from an image obtained by capturing an object other than a person moving in the room, and the second texture data is extracted by sequentially calculating the difference between the data captured in time series. By determining whether the second texture data matches or does not match, it is possible to accurately determine the presence or absence of a person, and to appropriately control blinking and dimming of the lighting fixture.
In addition, it is possible to easily specify an area for setting the first texture data from the image captured by the imaging device.

  Furthermore, the present invention includes the above lighting system in which the setting unit sets a plurality of the first texture data and groups them from images taken from different directions of the specific object.

  With this configuration, it is possible to easily specify an area for setting the first texture data from the image captured by the imaging device.

  In the illumination system according to the present invention, the determination unit may determine whether the first texture data and the second texture data are obtained by any one of a neural network, a fuzzy algorithm, and multiple regression analysis in pattern recognition. It also includes those that determine match or mismatch.

  With this configuration, even if the first texture data and the second texture data do not completely match, it is possible to accurately determine the presence or absence of a person, and to appropriately control blinking and dimming of the lighting fixture. It becomes possible.

  ADVANTAGE OF THE INVENTION According to this invention, the presence or absence of the person in lighting space can be determined correctly, and the lighting system which can control blinking and light control of a lighting fixture appropriately can be provided.

  Hereinafter, an illumination system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a schematic configuration of an illumination system according to an embodiment of the present invention.

  In FIG. 1, the lighting system according to the present embodiment is based on one or more lighting fixtures L1 to Ln installed in a room, an image pickup apparatus 1 including an image sensor for picking up the room, and the image data captured. It is the structure provided with the judgment apparatus 2 which judges the presence or absence of a person in, and the illumination control apparatus 3 which controls lighting of the lighting fixtures L1-Ln according to the presence or absence of a person.

  The imaging device 1 is integrally configured to include a solid-state imaging device such as a CCD or CMOS, an optical lens, and a signal processing circuit, and is installed on a ceiling or a wall surface of the room to sequentially image the room in time series for determination. Image data is sent to the device 2. As the optical lens, it is desirable to use a wide-angle lens or a fish-eye lens in order to image a wide range in the room. In addition, the solid-state imaging device may be for color or monochrome, and it is necessary to determine whether a person has entered the room even when the illumination is turned off. Therefore, a CCD imaging device having sensitivity up to the infrared region is suitable.

  The determination device 2 includes a setting unit 21 that sets first texture data to be canceled from the image data captured by the imaging device 1, a storage unit 22 that stores the captured image data and first texture data, and the imaging device 1. By sequentially calculating the difference between the image data picked up by the calculation unit 23 for extracting the second texture data moving in the room, the first texture data and the second texture data are collated, and the coincidence or mismatch By determining whether there is a person in the room, and a display unit 25 for displaying an image captured by the imaging apparatus 1.

  The setting unit 21 inputs image data captured by the imaging device 1 in accordance with an operation of a texture reading activation switch (not shown), and cancels the first texture data that is canceled from the image data sequentially captured by the imaging device 1 in time series. Set.

  The first texture data is image data obtained by imaging a moving object other than a person in the room. For example, an image obtained by imaging a fan that swings, a pet that moves around the room, a curtain that is shaken by wind blown from a window, and the like. Even if the data is included and moves, it should be judged that it is not a person.

  The storage unit 22 includes a RAM, EPROM, EEPROM, and flash memory that can rewrite data. The storage unit 22 sequentially captures image data of the room in time series by the imaging device 1 and the first texture that has been set. Data and grouped first texture data group, flow line trajectory data of people in the room, dimming data of the lighting bases L1 to Ln set in advance corresponding to the position and number of people in the room, etc. To do.

  The calculation unit 23 is constituted by a microcomputer system, sequentially reads out image data obtained by imaging the room from the storage unit 22, calculates a difference between them, and extracts second texture data moving in the room.

  The determination unit 24 is configured by a microcomputer system common to the calculation unit 23, and compares the first texture data read from the storage unit 22 with the second texture data extracted by the calculation unit 23, and matches or does not match. Is determined based on a predetermined algorithm to be described later, thereby determining the presence or absence of a person in the room.

  If the first texture data and the second texture data match, it is determined that the person moving in the room is not a person, and if they do not match, it is determined that the person is a person. Then, based on the presence / absence of a person as a result of these determinations and the position where there is a person, the corresponding light control data is read from the storage unit 22 and sent to the illumination control device 3.

  If it is determined that a person moves in the room, the trajectory is subsequently tracked, and the position is stored in the storage unit 22 when stopped.

  The display unit 25 is composed of, for example, a liquid crystal display having a touch panel function. The display unit 25 displays an image captured by the imaging device 1 and applies first texture data by applying pressure on the displayed image with a finger or a touch pen. Specifies the image area to be canceled for setting.

  The lighting control device 3 is configured by a microcomputer system common to the calculation unit 23, and inputs dimming data sent from the determination unit 24, and controls dimming and brightness of the lighting fixtures L1 to Ln. Generate a signal.

  Next, the operation of the determination device 2 in the lighting system of the present embodiment configured as described above will be described. FIG. 2 is a flowchart for explaining a processing procedure in the determination apparatus 2.

  First, the setting unit 21 determines whether or not a texture reading activation switch (not shown) for starting the setting of the first texture data has been pressed (step S11).

  If it is determined that the texture read switch activation switch has been pressed, a first texture data setting process subroutine is executed in step S12.

  On the other hand, if it is determined that the texture read switch activation switch has not been pressed, the processes after step S13 are executed using the first texture data already stored in the storage unit 22.

  In step S <b> 13, whether or not the second texture data moving in the room is extracted by sequentially calculating the indoor image data captured by the imaging device 1 from the storage unit 22 and calculating the difference between them. Determine.

  If the second texture data is extracted as a result of the determination, a subroutine of a process for determining whether or not the moving object is a person is executed in the next step S14.

  On the other hand, if the second texture data is not extracted in the procedure of step S13, the procedure returns to the procedure of step S11 after a predetermined time delay in the procedure of step S17.

  In step S15, based on the result of executing the subroutine of step S14, it is determined whether or not the person moving in the room is a person.

  As a result of the determination, if the moving object is a person, in the next step S16, a subroutine for performing lighting control such as blinking and dimming of the lighting fixtures L1 to Ln according to the number of people in the room, the position of the person, etc. Execute.

  On the other hand, if the person who moves in the procedure of step S15 is not a person, the procedure returns to the procedure of step S11 after a predetermined time delay in the procedure of step S17.

  Next, the first texture data setting process will be described. FIG. 3 is a flowchart for explaining the processing procedure of the first texture data setting processing subroutine shown in FIG.

  First, the setting unit 21 designates and inputs an image area to be canceled from an image captured in the room (step S131). The image area to be canceled is a specific object that is moving but not human, such as a small animal such as a moving pet, a fan that swings, a device that moves and rotates, a curtain that is swayed by the wind, etc. .

  Here, a method for designating an image area to be canceled from an image captured in a room will be described. 4 to 6 are explanatory diagrams for describing a method for designating an image area to be canceled from an image obtained by capturing an image of the room.

  Fans, pets, curtains, and the like have different forms, sizes, patterns, and the like, and therefore need to be set separately as the first texture data. Therefore, the image area to be canceled is individually designated.

  FIG. 4 is a diagram illustrating a method of designating an image area to be canceled for setting the first texture data, using the indoor captured image displayed on the display unit 25 that forms part of the determination device 2. .

  In FIG. 4, after pressing the texture reading start switch 26, an image area to be canceled, for example, an image area in which the image of the area is canceled by applying pressure by tracing the fan portion of the fan F with a finger or the touch pen 27. To the setting unit 21 to set the first texture data.

  5 and 6 are diagrams showing another method of designating an area to be canceled, and an object of an image area to be canceled is directly designated using a laser pointer.

  In FIG. 5, after pressing a texture reading start switch (not shown), a closed curve is drawn around the fan portion of the fan F, which is the object, using the laser pointer LP, and this is imaged by the imaging device 1 and its image data. Is set in the setting unit 21 and set as first texture data.

  Similarly, in FIG. 6, a closed curve is drawn with the laser pointer LP on the curtain C of the target object, which is an area to be canceled, and this is captured by the imaging device 1 and recognized by the setting unit 21 of the determination device 2. Set the texture data.

  The point indicated by the laser pointer that indicates the range of the object is a point, but on a continuous image, it is a closed curve with a series of points, so it is possible to easily extract the image area surrounded by the closed curve by calculation. is there.

  By such a method as described above, a fan, a pet, a curtain, or the like having a different shape, size, pattern, or the like in the room can be easily designated as a cancel image area for setting the first texture data.

  The laser pointer LP may be integrated with the slave unit of the determination device 2 or the remote control of the illumination control device 3.

  Returning to FIG. 3, cancel images i (i = 1 to n) are acquired at regular time intervals while the specified image area to be canceled is tracked and imaged by the imaging device 1 (step S <b> 132). Data Xij (j = 1 to m) is calculated from i (step S133).

  Next, in step S134, an image h (h =! To k) outside the cancel region is acquired, and data Xkj (j = 1 to m) is calculated from the acquired image h (step S135).

  From these data, the following equation is made simultaneous and As is obtained by multiple regression analysis (step S136).

1 = A1Xij + A2Xij + A3Xij + ... + AmXij
1 = A1Xij + A2Xij + A3Xij + ... + AmXij
1 = A1Xij + A2Xij + A3Xij + ... + AmXij
...
1 = A1Xkj + A2Xkj + A3Xkj + ... + AmXkj
1 = A1Xkj + A2Xkj + A3Xkj + ... + AmXkj
1 = A1Xkj + A2Xkj + A3Xkj + ... + AmXkj
...
However, i = 1 to n, j = 1 to m, and k = 1 to n.

  The object to be set as the first texture data, such as an electric fan, a pet, or a curtain, changes the image picked up by the image pickup device 1 depending on its position and state, so the first texture data stored in the storage unit 22 And not exactly match. Accordingly, it is necessary to perform a matching determination process by recognizing a pattern using a technique such as a neural network, a fuzzy algorithm, or multiple regression analysis.

  In addition, the accuracy of pattern recognition is improved by preparing a plurality of first texture data prepared by imaging from different angles for one of the objects whose images are to be canceled. For example, a fan that shakes the fan's head, a moving pet, and the like are sequentially imaged in time series, and a plurality of first texture data are set and grouped from the obtained individual images. The grouped first texture data is stored in the storage unit 22.

  FIG. 7 is an explanatory diagram showing a case where three first texture data to be grouped are set by imaging the fan in three states with different swing angles. U and V are orthogonal coordinates in the image data.

  Several parameters are extracted from the three image groups shown in FIG. Examples of parameters to be extracted include the following.

X1: Spatial frequency peak value in the U direction X2: Spatial frequency peak value in the V direction X3: Pixel area included in the outermost periphery extracted from the edge X4: Red area in the region of X3 X5: Blue area in the region of X3

  Similarly, the parameters X′1 to X′5 are extracted from an image that is not a cancel image.

  When determining by multivariate analysis, when X1 to X5 are substituted into the following equation (1), Y = 1 (is a cancel image), and when X′1 to X′5 are substituted, Y = 0 Judgment is made by formulating (it is a person) and solving this.

  Y = AX1 + BX2 + CX3 + DX4 + EX5 (1)

  Next, human judgment processing will be described. FIG. 8 is a flowchart for explaining the processing procedure of the human judgment processing subroutine shown in FIG.

  First, in step S151, the image data Y obtained by Expression (1) is acquired from the moving object of the image captured by the imaging device 1.

  Next, data Zj (j = 1 to k) is calculated from the acquired image Y (step S152).

In step S203, the following equation (2) is calculated.
P = Zj + A2Zj + A3Zj + A4Zj + ... + AmZj (2)
However, j = 1 to m.

  Subsequently, it is determined whether or not the value of P calculated by Expression (2) is larger than 0.5 (step S154). If the result is greater than 0.5, it is determined that the moving object of the image captured by the imaging device 1 is a person (step S155).

  On the other hand, if the value of P is smaller than 0.5 in the procedure of step S154, it is determined that the captured image is not a person (step S156).

  Next, an example of determining the coincidence of the first texture data using a neural network will be described. FIG. 9 is a diagram showing a neural network forming a pattern recognition circuit.

  In FIG. 9, a pattern matching determination algorithm is established by inputting a texture group to be canceled in X1 to X5 and parameters from a texture that is not an image to be canceled.

  As described above, according to the lighting system according to the embodiment of the present invention, it is possible to accurately determine the presence / absence of a person in a room, and when it is determined that there is a person, the number and position of the person are determined. Accordingly, it is possible to appropriately control blinking and dimming of the lighting fixture.

  The lighting system of the present invention can accurately determine the presence or absence of a person in the lighting space, and has the effect of being able to appropriately control the blinking and dimming of the lighting fixture. Etc. are useful.

The figure which shows schematic structure of the illumination system which concerns on embodiment of this invention. The flowchart for demonstrating the process sequence of a judgment apparatus in the illumination system which concerns on embodiment of this invention. The flowchart which shows the process sequence of the 1st texture data input process subroutine shown in FIG. 2 in the illumination system which concerns on embodiment of this invention. Explanatory drawing for demonstrating the method to designate the area | region to cancel from the image which imaged the room on the image displayed on the display part in the illumination system which concerns on embodiment of this invention. Explanatory drawing for demonstrating the method to designate the area | region canceled from the image which imaged the room | chamber interior using the laser pointer in the illumination system which concerns on embodiment of this invention. Explanatory drawing for demonstrating the method to designate the area | region canceled from the image which imaged the room | chamber interior using the laser pointer in the illumination system which concerns on embodiment of this invention. In the lighting system which concerns on embodiment of this invention, explanatory drawing which shows the case where a fan is imaged in three states from which a head swing angle differs, and the 3rd 1st texture data set to group is set. The flowchart for demonstrating the process sequence of the human judgment process subroutine shown in FIG. 2 in the illumination system which concerns on embodiment of this invention. The schematic diagram which forms a pattern recognition circuit using a neural network in the illumination system which concerns on embodiment of this invention The figure which shows the conventional method of detecting a person in a lighting system

Explanation of symbols

1 Imaging device (image sensor)
2 Judgment device 21 Setting unit 22 Storage unit 23 Calculation unit 24 Judgment unit 25 Display unit 3 Lighting control device L1 to Ln

Claims (3)

A lighting system that controls blinking and dimming of a lighting fixture according to the presence or absence of a person in a lighting space,
An imaging device that sequentially images the illumination space in time series and outputs image data;
A determination device for determining presence or absence of a person in the illumination space based on the image data;
A lighting control device that controls blinking and dimming of the lighting fixture according to a determination result of the determination device;
The determination device includes:
A setting unit for setting first texture data to be canceled from the image data;
A storage unit for storing the image data and the first texture data;
A calculation unit for extracting second texture data moving in the illumination space by sequentially calculating a difference between the image data in time series;
A determination unit for determining presence or absence of a person in the illumination space in accordance with a match or mismatch between the first texture data and the second texture data;
A display unit that displays an image captured by the imaging device,
The setting unit
An illumination system that sets the first texture data based on an image region that is specified for an image of the illumination space displayed on the display unit and includes a specific object that moves in the illumination space other than a person . The lighting system according to claim 1 ,
The setting unit
An illumination system that sets and groups a plurality of the first texture data from images taken from different directions of the specific object. The illumination system according to claim 1 or 2 ,
The determination unit
An illumination system that determines a match or mismatch between the first texture data and the second texture data by any one of a neural network, a fuzzy algorithm, and multiple regression analysis in pattern recognition.

Images