WO2020107404A1 - On-site analysis system for doorway image - Google Patents

Abstract

An on-site analysis system for a doorway image, comprising: a plurality of infrared detection devices (4), provided in a washroom, each infrared detection device (4) being provided above a corresponding urinal (2) for detecting whether there is an object in front of the corresponding urinal (2), and when there is an object, issuing the current use state and the serial number of the corresponding urinal (2); a footstep detection device, provided at a doorway position of the washroom for performing footstep detection on the surroundings, the footstep detection comprising: when recognizing a footstep sound from a person, detecting a change tendency of the footstep sound from the person over time; when the footstep sound is increasing, issuing a person approach signal; when the footstep sound is decreasing, issuing a person leaving signal; and when no footstep sound is recognized, not outputting any signal. The present system improves the smart control level of a urinal device.

Description

A porch image field analysis system Technical field

The invention relates to the field of image analysis, in particular to a porch image field analysis system.

Background technique

Image analysis is not limited to classifying specific areas in an image into a fixed number of categories, it mainly provides a description of the image being analyzed. To this end, it is necessary to use both pattern recognition technology and knowledge base on image content, that is, the content of knowledge expression in artificial intelligence. Image analysis needs to use image segmentation to extract the characteristics of the image, and then describe the image symbolically. This description can not only answer whether there is a specific object in the image, but also make a detailed description of the image content.

Summary of the invention

The present invention provides a porch image on-site analysis system, which firstly realizes the instantaneous detection of the passing crowd through the targeted passing crowd detection mechanism, and then, can process each urinal separately, based on the current In the use state, when there are people in the porch, it is decided whether to extend the shielding baffle, thereby improving the directional sight shielding ability of the urinal and protecting the safety and privacy of the urinal user.

According to an aspect of the present invention, a porch image on-site analysis system is provided, the system including:

Multiple infrared detection devices are installed in the toilet, and each infrared detection device is installed above the corresponding urinal to detect whether there is an object in front of the corresponding urinal, and when there is an object, the current use status is issued And the corresponding urinal number;

A footstep detection device, which is set at the porch position of the restroom, is used to detect footsteps around the footsteps. The footstep detection includes: when a person's footstep sound is recognized, detecting a change trend of the person's footstep sound over time, when the change trend is increasing , Send a signal of personnel approaching, when the change trend is reduced, send a signal of personnel leaving; when no sound of footsteps is recognized, no signal is output;

On-site video equipment, set at the porch of the toilet, connected to the footstep detection equipment;

Wherein, the on-site video recording device is used to switch from the power saving state to the working state when the personnel approach signal is received, and start image capture of the porch position to obtain and output multiple frames of the current porch within a preset time interval The image is also used to switch from the working state to the power saving state when the personnel leaving signal is received, and to stop the image capturing of the porch position;

A resolution detection device, connected to the on-site video recording device, is used to receive multiple frames of current porch images within a preset time interval, perform maximum resolution detection on each frame of current porch images to obtain and output multiple frames of current porch images Maximum resolution in

Customizing a noise reduction device, connected to the resolution detection device, for receiving the maximum resolution, determining a corresponding noise reduction amplitude based on the maximum resolution, and based on the determined noise reduction amplitude for the current multi-frame porch The image is subjected to noise reduction processing to obtain and output a corresponding multi-frame noise reduction processing image, where the larger the maximum resolution, the smaller the noise reduction amplitude.

It can be seen from this that the benefits of the present invention are:

1. A customized mechanism of noise reduction amplitude based on the maximum resolution of the image is adopted to improve the noise reduction effect of the image;

2. By analyzing the data of the multi-frame images within the preset time interval, determine the component with the largest variation, and based on the corresponding image of the determined component, perform the search operation of the target shape;

3. Each urinal can be processed separately to determine whether to extend the shielding baffle when there are people on the porch based on the current usage status of each urinal, thereby improving the directional sight shielding ability of the urinal.

BRIEF DESCRIPTION

In order to more clearly explain the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. Those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

The embodiments of the present invention will be described below with reference to the drawings, in which:

FIG. 1 is a schematic structural diagram of a toilet where a porch image field analysis system is shown according to an embodiment of the present invention.

detailed description

The embodiment of the porch image field analysis system of the present invention will be described in detail below with reference to the drawings.

For public toilets, the space is limited. In the limited space, bathroom products such as washbasins and toilets are all indispensable. At the same time, the furnishings must be methodical, which does not affect the individual use of each product. It takes some effort. In the design of public toilets, we must first pay attention to the size of various bathroom facilities. Not crowded is the major premise. It is necessary to ensure that each bathroom equipment has a certain distance in the use space. Secondly, the public toilet floor should be high outside and low inside. The outside can be used to put a washbasin and toilet, which can not only waterproof and moisture-proof, but also achieve the effect of separating wet and dry.

For public toilets used by men, multiple urinals are a necessary sanitary equipment. The relative position and arrangement of multiple urinals in each public toilet are often unsatisfactory. Therefore, people in the past can see the sight of each urinal. No, it is difficult to formulate an effective vision shielding plan.

In order to overcome the above deficiencies, the present invention builds a porch image field analysis system.

FIG. 1 is a schematic structural diagram of a toilet where a porch image field analysis system is shown according to an embodiment of the present invention, where 1 is a wall where multiple urinals are installed.

The porch image field analysis system according to the embodiment of the present invention includes:

A plurality of infrared detection devices 4 are provided in the toilet, and each infrared detection device is provided above the corresponding urinal 2 for detecting whether there is an object in front of the corresponding urinal, and when there is an object, the current Use status and the corresponding urinal number;

A footstep detection device, which is set at the porch position of the restroom, is used to detect footsteps around the footsteps. The footstep detection includes: when a person's footstep sound is recognized, detecting a change trend of the person's footstep sound over time, when the change trend is increasing , Send a signal of personnel approaching, when the change trend is decreasing, send a signal of personnel leaving; when no sound of footsteps is recognized, no signal is output.

Next, the specific structure of the porch image field analysis system of the present invention will be further explained.

The porch image field analysis system may further include:

On-site video equipment, set at the porch of the toilet, connected to the footstep detection equipment;

Wherein, the on-site video recording device is used to switch from the power saving state to the working state when the personnel approach signal is received, and start image capture of the porch position to obtain and output multiple frames of the current porch within a preset time interval The image is also used to switch from the working state to the power saving state when the personnel leaving signal is received, and to stop the image capturing of the porch position.

The porch image field analysis system may further include:

A resolution detection device, connected to the on-site video recording device, is used to receive multiple frames of current porch images within a preset time interval, perform maximum resolution detection on each frame of current porch images to obtain and output multiple frames of current porch images Maximum resolution in

Customizing a noise reduction device, connected to the resolution detection device, for receiving the maximum resolution, determining a corresponding noise reduction amplitude based on the maximum resolution, and based on the determined noise reduction amplitude for the current multi-frame porch Performing noise reduction processing on the image to obtain and output corresponding multi-frame noise reduction processing images, wherein the larger the maximum resolution, the smaller the noise reduction range;

A component data extraction device, connected to the resolution detection device, is used to receive multiple frames of noise reduction processing images, and perform the following processing on each frame of noise reduction processing images: parse out each pixel in the noise reduction processing images The Y component, U component, and V component determine the sum of the Y component of each pixel as the sum of the Y component of the noise reduction processing image, and determine the sum of the U component of each pixel as the U of the noise reduction processing image Component sum, and the sum of the V components of each pixel is determined as the V component sum of the noise reduction processing image;

A time-wise processing device, connected to the component data extraction device, for receiving the Y component sum of each frame noise reduction processing image, the U component sum of each frame noise reduction processing image, and the V component of each frame noise reduction processing image Component sum, and determine the change amplitude of the Y component sum of the multi-frame noise reduction processing image, the change amplitude of the U component sum of the multi-frame noise reduction processing image, and the change amplitude of the V component sum of the multi-frame noise reduction processing image. The largest component is output as the calibration component;

The image extraction device is used to receive the latest frame of noise reduction processing image, and is also connected to the time-wise processing device to receive the calibration component, based on the calibration component from the latest frame of noise reduction processing image acquisition and the The component image corresponding to the calibration component is searched for a pattern conforming to the standard shape of the person from the component image, and when a pattern conforming to the standard shape of the person is searched, a target determination signal is issued; otherwise, a target uncertainty signal is issued;

Multiple pairs of shielding baffles 3 are provided in the toilet, and each pair of shielding baffles are provided on both sides of the corresponding urinal 2 for contracting in the ceramic wall of the corresponding urinal in the default state and also for In the extended state, it is unfolded from the ceramic wall of the corresponding urinal to increase the length of both sides of the corresponding urinal to improve the sight shielding ability of the corresponding urinal;

The baffle drive device is respectively connected to the image extraction device, a plurality of infrared detection devices and a plurality of pairs of shielding baffles, and is used to control and urinate based on the received number of each urinal when receiving the target determination signal A pair of shielding baffles with the urinal in the current use state corresponding to the number of the bucket are deployed.

The porch image field analysis system may further include:

A timer is connected to the component data extraction device and is used to provide timing operation for the component data extraction device.

Meanwhile, the on-site analysis method of the porch image according to the embodiment of the present invention includes:

Use multiple infrared detection devices, installed in the toilet, each infrared detection device is located above the corresponding urinal, used to detect whether there is an object in front of the corresponding urinal, and when there is an object, issue the current use Status and corresponding urinal number;

Use a footstep detection device, which is set at the porch position of the toilet for footstep detection around the footsteps. The footstep detection includes: when a person's footstep sound is recognized, the change trend of the person's footstep sound over time is detected, and when the change trend is increased At this time, the personnel approach signal is issued. When the change trend is decreasing, the personnel departure signal is issued; when the sound of personnel footsteps is not recognized, no signal is output.

Next, the specific steps of the porch image on-site analysis method of the present invention will be further described.

The porch image field analysis method may further include:

Use on-site video equipment, set in the porch position of the toilet, and connect with the footstep detection equipment;

Wherein, the on-site video recording device is used to switch from the power saving state to the working state when the personnel approach signal is received, and start image capture of the porch position to obtain and output multiple frames of the current porch within a preset time interval The image is also used to switch from the working state to the power saving state when the personnel leaving signal is received, and to stop the image capturing of the porch position.

The porch image field analysis method may further include:

Use a resolution detection device, connected to the on-site video recording device, to receive multiple frames of current porch images within a preset time interval, perform the maximum resolution detection on each frame of current porch images to obtain and output multiple frames of current porch The maximum resolution in the image;

Use a custom noise reduction device, connected to the resolution detection device, for receiving the maximum resolution, determine a corresponding noise reduction amplitude based on the maximum resolution, and based on the determined noise reduction amplitude for the current multi-frame Perform noise reduction processing on the porch image to obtain and output corresponding multi-frame noise reduction processing images, where the larger the maximum resolution, the smaller the noise reduction range;

Use a component data extraction device, connected to the resolution detection device, to receive multiple frames of noise reduction processing images, perform the following processing on each frame of noise reduction processing images: resolve each pixel in the noise reduction processing images Y component, U component and V component, determine the sum of the Y component of each pixel as the sum of the Y component of the noise reduction process image, determine the sum of the U component of each pixel as the noise reduction process image The sum of U components and the sum of V components of each pixel are determined as the sum of V components of the noise reduction processing image;

Use a time-wise processing device, connected to the component data extraction device, for receiving the Y component sum of each frame noise reduction processing image, the U component sum of each frame noise reduction processing image, and the V component sum, and determine the change amplitude of the Y component sum of the multi-frame noise reduction process image, the U component sum of the multi frame noise reduction process image, and the V component sum of the multi frame noise reduction process image, will change The component with the largest amplitude is output as the calibration component;

Use an image extraction device for receiving the latest frame of noise reduction processing image, and also connect with the time-wise processing device to receive the calibration component, based on the calibration component from the latest frame of noise reduction processing image acquisition and The component image corresponding to the calibration component is searched for a pattern conforming to the standard shape of the person from the component image, and when a pattern conforming to the standard shape of the person is searched, a target determination signal is issued; otherwise, a target uncertainty signal is issued;

Use multiple pairs of shielding baffles, set in the toilet, each pair of shielding baffles are set on both sides of the corresponding urinal, used to shrink in the ceramic wall of the corresponding urinal in the default state, and also used in In the extended state, it is unfolded from the ceramic wall of the corresponding urinal to increase the length of both sides of the corresponding urinal to improve the sight shielding ability of the corresponding urinal;

Use a baffle drive device, respectively connected to the image extraction device, a plurality of infrared detection devices and a plurality of pairs of shielding baffles, used to control and based on the received number of each urinal when receiving the target determination signal A pair of shielding baffles with the current urinal corresponding to the number of the urinal are deployed.

The porch image field analysis method may further include:

Use a timer to connect with the component data extraction device and provide a timing operation for the component data extraction device.

In addition, multiple ultrasonic detection devices can also be used to replace the multiple infrared detection devices, the multiple ultrasonic detection devices are provided in the toilet, and each ultrasonic detection device is provided above the corresponding urinal for the corresponding To detect whether there is an object in front of the urinal.

Due to the strong directivity of ultrasonic waves, the slow energy consumption and the long distance of propagation in the medium, ultrasonic waves are often used for distance measurement, such as rangefinders and level measuring instruments. The use of ultrasonic testing is often relatively fast, convenient, simple calculation, easy to achieve real-time control, and can meet industrial requirements in terms of measurement accuracy.

In order to study and use ultrasound, many ultrasound generators have been designed and made. Generally speaking, ultrasonic generators can be divided into two categories: one is to generate ultrasonic waves by electrical means, and the other is to generate ultrasonic waves by mechanical means. The electrical methods include piezoelectric type, magnetostrictive type and electric type, etc.; the mechanical methods include Galtonian flute, liquid whistle and air flow whistle. The frequency, power and sound wave characteristics of the ultrasonic waves they produce are different, so their uses are also different. At present, the piezoelectric ultrasonic generator is more commonly used.

The piezoelectric ultrasonic generator actually uses the resonance of the piezoelectric crystal to work. There are two piezoelectric wafers and a resonance plate inside the ultrasonic generator. When a pulse signal is applied to his poles, and its frequency is equal to the natural oscillation frequency of the piezoelectric wafer, the piezoelectric wafer will resonate and drive the resonance plate to vibrate, which generates ultrasonic waves. Conversely, if no voltage is applied between the two electrodes, when the resonance plate receives ultrasonic waves, it will press the piezoelectric wafer to vibrate and convert mechanical energy into electrical signals. At this time, he becomes an ultrasonic receiver.

The porch image field analysis system of the present invention is used to deal with the technical problems of different layout methods of public toilet urinals in the prior art that affect the design of the visual field shielding scheme. Each urinal is processed separately to be based on the current status of each urinal. In the use state, when there are people in the porch, decide whether to extend the shielding baffle. More importantly, in the detection mode for the porch personnel, a custom mechanism based on the maximum resolution of the image noise reduction amplitude is adopted, and the preset The data analysis of the multi-frame images within the time interval determines the component with the largest change range, and performs the search operation of the target shape based on the corresponding image of the determined component, thereby solving the above technical problem.

It can be understood that although the present invention has been disclosed as the preferred embodiments above, the above embodiments are not intended to limit the present invention. For anyone skilled in the art, without departing from the scope of the technical solutions of the present invention, the technical contents disclosed above can be used to make many possible changes and modifications to the technical solutions of the present invention, or to be modified to equivalent changes, etc.实实施例。 Effective embodiments. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims (4)

1. A porch image field analysis system, characterized in that the system includes:

   Infrared detection devices are installed in the toilets. Each infrared detection device is installed above the corresponding urinal. It is used to detect whether there is an object in front of the corresponding urinal. When there is an object, the current use status is issued. And the corresponding urinal number;

   A footstep detection device, which is set at the porch position of the restroom, is used to detect footsteps around the footsteps. The footstep detection includes: when a person's footstep sound is recognized, detecting a change trend of the person's footstep sound over time, when the change trend is increasing , Send a signal of personnel approaching, when the change trend is decreasing, send a signal of personnel leaving; when no sound of footsteps is recognized, no signal is output.
2. The on-site analysis system for porch images according to claim 1, further comprising:

   On-site video equipment, set at the porch of the toilet, connected to the footstep detection equipment;
3. The porch image field analysis system according to claim 2, further comprising:

   A resolution detection device, connected to the on-site video recording device, is used to receive multiple frames of current porch images within a preset time interval, perform maximum resolution detection on each frame of current porch images to obtain and output multiple frames of current porch images Maximum resolution in

   Customizing a noise reduction device, connected to the resolution detection device, for receiving the maximum resolution, determining a corresponding noise reduction amplitude based on the maximum resolution, and based on the determined noise reduction amplitude for the current multi-frame porch Performing noise reduction processing on the image to obtain and output corresponding multi-frame noise reduction processing images, wherein the larger the maximum resolution, the smaller the noise reduction range;

   A component data extraction device, connected to the resolution detection device, is used to receive multiple frames of noise reduction processing images, and perform the following processing on each frame of noise reduction processing images: parse out each pixel in the noise reduction processing images The Y component, U component, and V component determine the sum of the Y component of each pixel as the sum of the Y component of the noise reduction processing image, and determine the sum of the U component of each pixel as the U of the noise reduction processing image Component sum, and the sum of the V components of each pixel is determined as the V component sum of the noise reduction processing image;

   A time-wise processing device, connected to the component data extraction device, for receiving the Y component sum of each frame noise reduction processing image, the U component sum of each frame noise reduction processing image, and the V component of each frame noise reduction processing image Component sum, and determine the change amplitude of the Y component sum of the multi-frame noise reduction processing image, the change amplitude of the U component sum of the multi-frame noise reduction processing image, and the change amplitude of the V component sum of the multi-frame noise reduction processing image. The largest component is output as the calibration component;

   The image extraction device is used to receive the latest frame of noise reduction processing image, and is also connected to the time-wise processing device to receive the calibration component, based on the calibration component from the latest frame of noise reduction processing image acquisition and the The component image corresponding to the calibration component is searched for a pattern conforming to the standard shape of the person from the component image, and when a pattern conforming to the standard shape of the person is searched, a target determination signal is issued; otherwise, a target uncertainty signal is issued;

   The shielding baffle is mostly used in the toilet. Each pair of shielding baffle is set on both sides of the corresponding urinal. It is used to shrink in the ceramic wall of the corresponding urinal in the default state and also used in the extended state. Next, unfold from the ceramic wall of the corresponding urinal to increase the length of both sides of the corresponding urinal to improve the sight shielding ability of the corresponding urinal;

   The baffle drive device is respectively connected to the image extraction device, a plurality of infrared detection devices and a plurality of pairs of shielding baffles, and is used to control and urinate based on the received number of each urinal when receiving the target determination signal A pair of shielding baffles with the urinal in the current use state corresponding to the number of the bucket are deployed.
4. The on-site porch image analysis system according to claim 3, further comprising:

   A timer is connected to the component data extraction device and is used to provide timing operation for the component data extraction device.

Images