JPH0652535B2 - Setting method for binarization level in traffic detector - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a binarization level setting method in a passer-by-passage detecting device for detecting the number of people entering and leaving a department store, an exhibition hall, an exhibition hall, or the like by using a television camera image. Is.

[0002]

2. Description of the Related Art Conventionally, as a method of counting the number of pedestrians who enter and exit through a doorway of a department store, an exhibition hall, etc., a method using a hand-counting counter is used, or a transmitter / receiver facing each other is used. A method using an optical means for detecting the interruption of the light beam between them, a method using ultrasonic waves with a plurality of sets of ultrasonic transmitters and receivers, or a mat switch that automatically counts the number of people stepped on Methods etc. are known. But,
It is difficult for these passerby number detection devices to count accurately in a congested state (for example, a large number of people pass by at once), when some people are standing still, or when a person with a large physique passes by. is there.

Therefore, in order to eliminate the above defects,
There has been proposed a passer-by-passage number detection apparatus which processes an image signal of a television camera to detect the moving direction and the number of persons (Japanese Patent Application No. 3-164076 presented by the applicant of the present application). This converts the signal from the TV camera into a digital signal, RA
An image input section to be written in M (write / read memory), a signal processing section for performing arithmetic processing of this image data according to a program stored in a ROM (read only memory), and a result of arithmetic processing by the signal processing section are displayed. The signal processing unit is configured by a display unit, performs fixed difference processing for binarizing a difference between two images of a background image and a sample image and compressing the number of images, and a difference pixel obtained by the fixed difference. It is a device for detecting the number of passers-by, which calculates the moving direction and the number of people by counting and discriminating the number and the position.

[0004]

However, the binarization level of the fixed difference processing performed in the signal processing unit of the number-of-passengers detection apparatus is as follows:
Since it is handled as an absolute value (for example, the binarization level is 10
In that case, if the unnecessary portion is removed by the fixed difference processing on the + (pale) side 10 and the- (dark) side 10 based on the background gradation, there is a disadvantage that even the necessary portion disappears. For example, when there is a shadow of a person passing on the background, it is necessary to increase the binarization level in order to eliminate the shadow, but it is necessary to increase the binarization level. In the case, based on the background gradation, +
An image in the (light) shade of gray (for example, a person's clothes)
However, the number of pixels for one person could not be obtained, and the number of passersby could not be accurately counted.

In view of the above situation, the present invention sets the binarization level for each of the difference between the background image and the sample image to be positive or negative, thereby eliminating the above-mentioned problems. It is intended to provide a method of setting the activation level.

[0006]

The present invention comprises an image input section, a signal processing section and a display section, wherein the signal processing section binarizes the difference between the background image and the sample image to determine the number of pixels. In a passer-by-passage detecting device that calculates the moving direction and the number of people by performing fixed difference processing for compression and counting and determining the obtained difference pixel number and position, the binarization level in the fixed difference processing is set to the background image. In the sample image, the binarization level is individually set for the grayscale level above the standard and the grayscale level below the standard, based on the grayscale level of.

[0007]

As described above, the first image data is used as the background (capture of the background image), and the difference with the sample image captured from the next is taken to perform the binarization and the pixel number compression processing. After processing, the number of compressed pixels with fixed difference is counted, + and-signs are given depending on the pixel position, and the number of pixels changed from + direction to-direction and the number of pixels changed from-direction to + direction. Total the numbers separately. In this case, the binarization level for the pixels of the grayscale of the sample image that is larger (lighter) than the grayscale of each pixel of the captured background image is small. Can be removed.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the structure of a passer-by number detecting apparatus related to the binarization level setting method of the present invention, in which 1 is a television camera and 2 is a video analog signal from the television camera 1 which is a digital signal. The image input unit 3 which converts the image data into the image memory (not shown) and writes the image data in the image memory (not shown) reads the video data from the television camera 1 stored by the image input unit 2 and stores it in the program memory (not shown). A signal processing unit for performing a detection processing calculation according to a processing program, a display device 4 for displaying the result of the calculation processing by the signal processing unit 3, and a printing device 5 for printing the contents displayed on the display device 4. .

The operation of the signal processing unit 3 according to the present invention will be described below with reference to the flow chart of the detection processing program shown in FIG.

First, this flow chart shows the image input unit 2
The processing procedure for obtaining the moving direction and the number of people of the passersby is shown by using the image data captured by. Fixed difference processing that performs binarization and compression processing is performed by taking the difference between the first image data as the background and the sample image captured next. After that, the number of compressed pixels with a fixed difference is counted, + and-signs are given depending on the pixel position, and the number of pixels changed from the + direction to the − direction and the number of pixels changed from the − direction to the + direction are distinguished. Total and divide by the number of standard pixels for one person.

In the binarization processing in the processing operations, the difference between the two screens of the background image, which is a grayscale image, and the sample image is calculated, and the difference between the two screens and the binarization level are compared to obtain a large value. For example, if it is smaller than 1, it is binarized to 0 so that the person image can be handled only by calculating the number of pixels. Here, when the difference binarization processing with the sample image is performed with the background gray scale as the reference, the gray scale of the sample image larger than the reference and the gray scale of the sample image smaller than the reference are used to individually set the binarization level. Set. Specifically, the binarization level for the pixels of the sample image having a grayscale that is larger (lighter) than the background grayscale is small, and the opposite is large, so that only the shadow is left while the person is left and the binary difference is removed. Can be processed.

For example, in the case where a person and its shadow are present in the sample image, the binarization process for eliminating the shadow will be described. Since the shadow has a smaller gradation (darker) than the background gradation. If the binarization level for the pixel of the sample image having a smaller (darker) shade than the reference is set to a large value (for example, the binarization level 20), the shadow is subjected to the difference binarization process. Can be erased by. On the other hand, the binarization level for the pixels of the sample image having a gray scale that is larger (lighter) than the reference may be set to be small within a range not affected by the image noise (for example, the binarization level 10). In addition, since the grayscale of a person's head is usually smaller (darker) than the shade of a shadow for black hair, the binarization level for a sample image with a grayscale smaller than the background grayscale (dark) Is set between the gray scale of black hair and the gray scale of shadow, a person's clothing and head can be obtained as a differential binary image. This relationship is shown in FIG.

As described above, the number of difference pixels of the obtained person is divided by the reference number of pixels to obtain the number of persons. At this time, the reference number of pixels is within the detection area of the person at the upper limit of the standard physique distribution. Maximum difference pixel number and 1 / 1.5 of the difference maximum pixel number
By setting it as a value between 2 and 3, the error is eliminated depending on the physique of the passing person (large person or small person), even if a large person passes within the standard physique distribution range, it is not counted as two people, and Even if two small people pass, they are not counted as one person.

The above-described processing is sequentially processed for one sample image, and when the processing is completed for all the sample images, the processing for the next sample image is performed. If there is no difference pixel in the sample image, the above process is not performed and the process for the next sample image is performed.

Next, the flowchart of the processing program for background update and illuminance correction shown in FIG. 4 will be described. This flowchart of processing program form part of the background image acquisition process in the flowchart of the processing program for determining the moving direction and the number of passersby that shown in FIG. 2, n ₁ times the fixed difference processing shown in FIG. 2 If the difference pixel does not change,
Update the background image with a new sample image. Next, the binarization level when performing difference processing with the pixel of the sample image for the pixel based on the grayscale of each pixel of the background image that has been captured is set in advance. Set by. After setting the binarization level for each pixel, the sample image capturing process shown in FIG. 2 is performed. Further, when the difference pixels generated by the fixed difference processing shown in FIG. 2 are not eliminated by consecutive sample image acquisition of n ₂ times, the process shifts to n ₁ times pixel number change determination processing.

[0017]

As described above, in the binarization level setting method in the number-of-passengers detection apparatus of the present invention, the binarization level is classified into positive and negative differences in the difference between the background and the sample image, and the sample image In the gray scale, the smaller (darker) binarization level is set larger than the gray scale of the background image, so that only shadows can be removed while leaving the person, and it is possible to detect the number of passersby. It will not cause an error.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a number-of-passengers detection apparatus according to the present invention.

FIG. 2 is a flowchart of a detection block of the signal processing circuit according to the embodiment.

FIG. 3 is an explanatory diagram showing a relationship between a gray scale corresponding to a person and a shadow and a binarization level.

FIG. 4 is a flowchart of a background update and illuminance correction program of the signal processing circuit according to the embodiment.

[Explanation of symbols]

 1 TV camera 2 image input unit 3 signal processing unit 4 display device 5 printing device

Claims (1)

[Claims] 1. An image input unit, a signal processing unit, and a display unit, wherein the signal processing unit performs fixed difference processing for binarizing a difference between a background image and a sample image and compressing the number of pixels, In a device for detecting the number of passers-by, which counts and discriminates the number and position of obtained difference pixels, the number of passersby detecting the binarization level in the fixed difference process in the sample image based on the background gradation. , A binarization level setting method in a passer-by-passenger number detection device, characterized in that the binarization level is individually set for a gray scale above the standard and a gray scale below the standard.