JPH03195922A - Detecting apparatus for degree of congestion - Google Patents

Abstract

PURPOSE:To enable highly-precise detection of the degree of congestion by dividing an image pickup screen into a plurality and by setting a criterion of detection for each divided screen. CONSTITUTION:When a zone A to a zone D are set, beginning with the nearest to a TV camera 1, on the occasion of setting a criterion of detection, the number of pixels for detecting one person is determined in such a manner that eight pixels are allocated to the zone A, six pixels to the zone B, four pixels for the zone C and two pixels to the zone D, for instance. First the content of a video memory on the occasion when there is no waiting person in an elevator hall is read in CPU 18A and stored in a reference data area of a data memory 18C. Next, fresh video data are read in the CPU 18A from the memory 14. Reference data and the fresh data are compared with each other for each pixel, and the pixels wherein a difference in brightness is a prescribed value or above are extracted and written in places corresponding to them respectively in a detected data area of the memory 18C so as to be stored. Now, when there is a waiting person, the criterion of detection is selected in accordance with the position on a screen and the degree of congestion is calculated. The calculated data thus obtained are outputted as data 13a on the degree of congestion to a group control device 17 through an output port 18E.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a device that detects the degree of crowding by processing data captured using a television camera or the like.

[Prior Art] FIG. 5 is a block diagram showing a conventional congestion level detection device disclosed in, for example, Japanese Patent Publication No. 57-30782.

In the figure, (1) is a television camera (hereinafter referred to as television camera) that images the elevator hall, and (2) is an analog-to-digital converter (hereinafter referred to as A/ (3) is an arithmetic unit that processes the digital signal and extracts the features of each pixel; (4) (5) is an arithmetic unit (3)
Memory that stores the output of (6) is memory (4) (5)
A comparator (7) is a counter that counts the number of mismatched pixel signals (described later) output from the comparator (6).

A conventional congestion level detection device is configured as described above, and an image of waiting passengers in an elevator hall captured by a television camera (1) is processed by a computing unit (3) to extract features in pixel units. . The processing results of a computing unit (3) when there are no waiting customers in the elevator hall are stored in a memory (4), and the processing results when there are waiting customers in the elevator hall are also stored in a memory (5). Then, the contents stored in the memories (4) and (5) are compared for each pixel by a comparator (6), and if the signals of corresponding pixels are different, a mismatched pixel signal is output. This signal is counted by a counter (7), and the counted value corresponds to the degree of congestion at that time.

[Problems to be Solved by the Invention] In the conventional congestion level detection device as described above, the congestion level is calculated based on the results of counting mismatched pixel signals. Camera (1
) will appear larger, and the number of pixels to detect - people will increase. Conversely, a person who is far from the television camera (1) appears small, and the number of pixels for detecting a person becomes smaller. Therefore 1
There is a problem in that the obtained degree of crowding varies greatly depending on the location of the person, and the detection accuracy inevitably becomes low.

This invention was made to solve the above-mentioned problems, and it is an object of the present invention to provide a crowding degree detection device that can detect the degree of crowding with high accuracy regardless of the distance between the imaging device and the person.

[Means for Solving the Problems] A congestion degree detection device according to the present invention divides an imaging screen into a plurality of parts, and sets a detection standard for each divided screen.

[Operation] In this invention, since the detection standard is set for each split screen, the number of pixels for detecting a person is set for each split screen, and the degree of congestion is detected based on this.

[Example] Figures 1 to 4 are diagrams showing an example of the present invention.
2 is a block diagram of main parts, FIG. 3 is a flowchart showing the operation, and FIG. 4 is an explanatory diagram of screen division. Portions similar to those of the conventional device are designated by the same reference numerals.

In Figure 1, (11) is an elevator hall, (12) is a waiting customer, (13) is a congestion level detection device, and an A/D converter (2
), and a video memory () that sequentially stores the imaging data digitized by the A/D converter (2) in synchronization with a video synchronization signal (not shown) and always stores the latest imaging data.
14), a detection standard setting means (15) that divides the screen and sets a detection standard for each divided screen, and calculates the number of waiting customers corresponding to the screen position based on the detection standard and generates congestion level data (13a). and a congestion degree calculation means (16) for calculating the congestion degree calculation means (16). (17) is an elevator group management device that manages elevators in groups based on congestion degree data (13a).

In Figure 2, (18) is a microcomputer (hereinafter referred to as microcomputer), which includes a CPU (18A), a program memory (18B), a data memory (18C), an input port (180), and an output port (18E). These are connected by a pass line (18F), the input port (1811) is connected to the video memory (14), and the output port (18B) is connected to the group management device (17).

Next, the operation of this embodiment will be explained with reference to FIGS. 3 and 4. The program shown in the flowchart of FIG. 3 is stored in the program memory (18B) of the microcomputer (18).

First, step (21) takes the elevator to the hall (11).
The contents of the video memory (14) when there is no waiting customer (12) in the CPU (1
8A). In step (22), CPU (1
The data read into the data memory (18C) is sequentially transferred to the data memory (18C).
) is stored in the reference data area. In step (23), new video data is transferred from the video memory (14) to the CP.
Load into U (18A). In step (24), the reference data and new data are compared pixel by pixel, and pixels with a difference in brightness greater than or equal to a predetermined value are extracted and placed in the location corresponding to each pixel in the detected data area of the data memory (18C). Write it down and memorize it.

Next, in step (25), it is determined from the data obtained from step (24) whether or not there is a waiting customer (12), and if there is not, the process returns to step (23), and step (23)
- Repeat (25). If there is a waiting customer (12), proceed to step (26), select a detection criterion to determine one or more people for each group of pixels according to the waiting customer (12), that is, the position of different pixels on the screen, and Calculate the number of customers waiting for each gathering. Then, in step (27), the degree of congestion of the entire elevator hall (11) is calculated based on the data obtained in step (26). In step (28), the calculated data is output to the group management device (17) as congestion degree data (13a) via the output port (18E).

By repeating the same steps below.

The degree of congestion is detected.

FIG. 4 illustrates an example of the size of a person depending on the position on the screen, as an example of the means for establishing a detection standard performed in step (26). Figure 4 shows an example in which the screen is divided into four vertically, and if zones A to D are defined from the one closest to the TV camera (1), the number of pixels for detecting a person is, for example, 8 pixels for zone A, Detection standards are set such as 6 pixels for the B zone, 4 pixels for the C zone, and 2 pixels for the D zone.

In addition, when dividing the screen, 2 or 4 or more divisions,
Alternatively, it is also possible to divide it in the left and right direction.

Furthermore, in the embodiment described above, the degree of congestion in the elevator hall (11) is detected.

For example, it can be applied to detect the degree of congestion on a station platform.

[Effect of the invention] As explained above, in this invention, the imaging screen is divided into 1
Since the detection standard is set for each split screen, the number of pixels for detecting a person is set for each split screen, and the detection error due to the position of one screen is corrected, making it possible to detect the degree of congestion with high accuracy. .

[Brief explanation of drawings]

1 to 4 are diagrams showing one embodiment of the congestion degree detection device according to the present invention, in which FIG. 1 is an overall configuration diagram, FIG. 2 is a block diagram of main parts, and FIG. 3 is a flowchart showing the operation. FIG. 4 is an explanatory diagram of screen division, and FIG. 5 is a block diagram showing a conventional congestion degree detection device. In the figure, (1) is an imaging device (TV camera), (11) is a place where people gather (elevator hall), (12) is a person (waiting customers), (13) is a congestion level detection device, (15) ) is a detection standard setting means, and (16) is a congestion degree calculation means. Note that the same reference numerals in the figures indicate the same parts.

Claims (1)

[Claims] In a device that captures an image of a place where people gather with an imaging device, and performs image processing on the image signal corresponding to this image capture screen to detect the degree of congestion of the place, the image capture screen is divided into a plurality of parts and each divided screen is divided into two. A congestion level detection device comprising a detection standard setting means for setting a detection standard.