Missed risk monitoring system for medical waste cold stores

Abstract

The present application relates to the field of image processing, in particular to a missing risk monitoring system for medical waste refrigerators, comprising a video data acquisition module, a light quality evaluation module, a sensitive factor acquisition module, a gray compensation module and a risk detection module. According to the monitoring video when transferring medical waste, each gray image and template image are obtained; according to the structural richness of each gray image and the average gray value of each differential gray level, the light quality evaluation index of each gray image is obtained; according to the light quality evaluation index of each gray image and adjacent gray image, the sensitive factor of each gray image is obtained, the gray compensation is carried out on each sensitive factor, and the missing risk of each gray image after the gray compensation is completed is detected. The present application can restore the details lost in the image due to light changes, ensure the integrity of the image detail information, and improve the accuracy of missing risk identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing, in particular to a missing risk monitoring system for medical waste refrigeration. BACKGROUND

[0002] The medical waste such as needle, infusion tube, gauze with blood or virus, if not strictly controlled and flows to the society, will be illegally used, and its danger is self-evident. The national dangerous waste list lists it as the first type of dangerous waste. Therefore, the daily generated medical waste will be handed over to the third party medical waste treatment company for professional destruction treatment. The medical waste refrigeration as a waste temporary storage place can maximize the inhibition of the reproduction and spread of viruses and bacteria in the waste, and the leakage of radioactive substances. The transportation process has a certain degree of danger, which is generally completed by professional medical personnel, but as long as human operation, there is always a security risk, so a monitoring system needs to be installed in the path of transportation to the refrigeration to monitor and alarm the risk operation of the staff, avoid the medical waste overturning and missing event, and also can control the danger of non-related personnel entering the refrigeration to cause the medical waste to flow out.

[0003] The monitoring system before the medical waste refrigeration needs to clearly identify whether the protective clothing of the staff is standard, whether the waste is missing and other detailed information, so the requirement for the monitoring imaging quality is high. However, due to the low temperature around the refrigeration, the mercury pressure in the energy-saving lamp is low, and the electronic components are unevenly heated and cooled, so that the light emitted by the energy-saving lamp is dark. The conventional video monitoring imaging is limited by the environmental light, and the imaging quality will be greatly reduced and the detailed image information will be easily lost in the low illumination environment. The traditional image preprocessing method can only realize brightness adjustment and contrast enhancement, but for the case that the image details have been lost, adjusting the basic parameters of a single video image has little effect on ensuring the integrity of image information, that is, the traditional image preprocessing method cannot guarantee the integrity of image information for the image whose details have been lost, and further leads to the omission and misjudgment of the risk event in the medical waste transportation process when the risk analysis is carried out according to the monitoring video image. SUMMARY

[0004] The present application provides a missing risk monitoring system for medical waste refrigeration to solve the existing problems.

[0005] The missing risk monitoring system for medical waste refrigeration of the present application adopts the following technical scheme:

[0006] An embodiment of the present application provides a missing risk monitoring system for medical waste refrigeration, which comprises the following modules: a video data acquisition module, a light quality evaluation module, a sensitive factor acquisition module, a gray compensation module and a risk detection module.

[0007] a video data acquisition module, which acquires monitoring video when medical staff transports medical waste, obtains continuous gray images, and takes the gray images when there is no one as template images;

[0008] a light quality evaluation module, which obtains differential gray levels of each gray image according to a gray histogram of each gray image and a gray histogram of the template image, and obtains a light quality evaluation index of each gray image according to average gray values of each differential gray level in each gray image and structural richness in the gray image;

[0009] a sensitive factor acquisition module, which acquires a target frame in each gray image, obtains an ideal gray level set in the target frame in each gray image according to light quality evaluation indexes of each gray image and adjacent gray images, and obtains a sensitive factor of each gray image according to the ideal gray level set corresponding to each gray image and a gray level set in a target frame of an adjacent gray image;

[0010] a gray compensation module, which compensates gray values of each sensitive factor according to the gray values of each sensitive factor of each gray image and gray values of pixel points at corresponding positions in adjacent gray images of each gray image, obtains gray values of each sensitive factor after gray compensation, and sequentially compensates gray values of each gray image;

[0011] a risk detection module, which detects each gray image after gray compensation for missing risk.

[0012] Preferably, the method for obtaining the differential gray levels of each gray image is as follows:

[0013] subtract the number of pixel points corresponding to each gray level in the gray histogram of each gray image from the number of pixel points corresponding to each gray level in the gray histogram of the template image, take an absolute value of the obtained difference value, when the absolute value is not 0, the gray level is a differential gray level, otherwise the gray level is not a differential gray level.

[0014] Preferably, the method for obtaining the structural richness in each gray image is as follows:

[0015] obtain a frequency of each differential gray level according to a ratio between the number of pixel points corresponding to each differential gray level in each gray image and the number of pixel points corresponding to all differential gray levels, calculate an information entropy according to the frequency of each differential gray level, and take the information entropy as the structural richness of each gray image.

[0016] Preferably, the method for obtaining the light quality evaluation index of each gray image is as follows:

[0017] The average gray value of each differential gray level of each gray image is normalized with the structural richness of the gray image, and the average of the normalized average gray value and the structural richness is taken as the light quality evaluation index of the gray image.

[0018] Preferably, the method for obtaining the set of ideal gray levels in the target frame of each gray image is as follows:

[0019] The ratio between the light quality evaluation index of each gray image and the light quality evaluation index of the adjacent gray image corresponding to the gray image is taken as the light quality change degree of the gray image, the light quality change degree is multiplied by each gray level contained in the target frame of the gray image, and the result is rounded down to obtain each ideal gray level of the gray image. The ideal gray levels constitute the set of ideal gray levels of the gray image.

[0020] Preferably, the method for obtaining the sensitive factor of each gray image is as follows:

[0021] The absolute value of the difference between each ideal gray level corresponding to each gray image and each gray level in the adjacent gray image of the gray image is calculated, and when the absolute value is less than or equal to 5, the gray level is a disappearing gray level of the adjacent gray image; otherwise, the gray level is not a disappearing gray level of the adjacent gray image.

[0022] Each pixel point corresponding to each disappearing gray level in the adjacent gray image at the corresponding position in each gray image is recorded as each sensitive factor of the gray image.

[0023] Preferably, the method for obtaining the gray value of the sensitive factor after gray compensation is as follows:

[0024] The difference between 1.0 and the hyperparameter is set as the reference weight of the gray value of each sensitive factor of each gray image, and the hyperparameter is set as the reference weight of the gray value of the corresponding pixel point in the adjacent gray image of each sensitive factor in each gray image. The gray value of the sensitive factor is weighted and summed with the gray value of the corresponding pixel point, and the result is taken as the gray value of the sensitive factor after gray compensation.

[0025] The beneficial effects of the present application are: first, the light quality of different images of the monitoring video image is calculated, that is, the light quality evaluation index; the light quality of each gray image is restored to the corresponding light quality at the previous moment through the light quality evaluation index of each gray image and its adjacent gray image, the ideal gray level corresponding to each gray level in the target frame of each gray image under the condition of not being disturbed by the change of light quality is obtained, and further, according to the difference between the ideal gray level and each gray level in the target frame corresponding to the actual previous moment, the gray level disappeared in the previous moment due to the change of light quality is obtained, so that the sensitive factor easily lost in each gray image due to the change of light is obtained, and finally the gray value of the sensitive factor of each gray image is weighted with the gray value of the pixel point at the corresponding position in the previous moment, so that the gray compensation of the dynamic target image in the uneven illumination scene is realized, the lost detail information in the image is restored, the integrity of the image detail information is ensured, and the image quality is improved. Compared with the image preprocessing method of the traditional monitoring system, the present method can greatly restore the easily lost detail information in the image, and compared with the similar compensation algorithm such as interpolation, the compensation effect obtained by the present method is more accurate, the integrity and reliability of the monitoring video information are improved, and the result obtained when detecting the security, omission, illegal operation and other risks in the transportation process of medical waste is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0027] Figure 1 The structure block diagram of the medical waste cold storage missing risk monitoring system of the present application. DETAILED DESCRIPTION

[0028] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific implementation, structure, features and effects of the medical waste cold storage missing risk monitoring system according to the present application will be described in detail as follows by combining with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0030] The specific scheme of the missing risk monitoring system for the medical waste cold storage provided by the present application will be specifically explained below in combination with the drawings.

[0031] Please refer to Figure 1 which shows the step flow chart of the missing risk monitoring system for the medical waste cold storage provided by an embodiment of the present application, and the method comprises the following steps:

[0032] The video data acquisition module is used to acquire the monitoring video when the medical staff transports the medical waste, and obtain continuous gray images.

[0033] The monitoring system cannot be directly installed in the cold storage and is generally installed on the path leading to the cold storage door. In order to more clearly capture the working process of the medical staff, the video monitoring installation height will not be too high. However, the illumination environment near the light source and far from the light source is different, and the image information collected by the monitoring system under low illumination conditions will have the phenomenon of lower local gray value, which will cause the details in this local area to be unable to be obtained, such as being unable to identify the face information of the miscellaneous personnel and being unable to accurately detect the missing of the medical waste. The high-illumination area and the low-illumination area in the monitoring video picture exist at the same time, and when the medical staff pushes the medical waste storage barrel from the high-illumination area to the low-illumination area, the internal information of the area corresponding to the medical staff will change. With the medical staff gradually moving away from the light source, the overall gray value of the area corresponding to the medical staff in the image becomes lower, and the internal information of the area gradually becomes blurred or even lost.

[0034] In order to better obtain the light environment of the medical staff at different moments in the monitoring video in the subsequent process, the present embodiment first selects the monitoring image when no one as a template image; then, the continuous images in the process of transporting the medical waste to the cold storage by the medical staff are intercepted in the memory of the monitoring system, and the template image and the continuous images in the transportation process are subjected to gray processing respectively, thereby obtaining continuous gray images.

[0035] The light quality evaluation module is used to evaluate the light quality of each gray image corresponding to the monitoring video, thereby obtaining the light quality evaluation index of each gray image.

[0036] The gray histogram corresponding to each gray image is obtained, the horizontal axis of the gray histogram represents each gray level, and the vertical axis represents the number of pixel points corresponding to each gray level. Similarly, the gray histogram corresponding to the template image is obtained;

[0037] Since the whole environment background is invariable, i.e. when there is no moving object in the monitoring video, i.e. there is no medical staff, the various gray levels in the background environment and the number of pixel points corresponding to the various gray levels are stable, therefore when there is a differential gray level in a gray image, it indicates that there is a moving object in the gray image, and the movement of the moving object causes the number of pixel points corresponding to the various gray levels in the gray image to change, i.e. a differential gray level appears. The gray histogram corresponding to each gray image and the gray histogram of the template image are respectively differentiated, for example, for the gray histogram corresponding to the s-th gray image, the absolute value of the difference between the number of pixel points of the k-th gray level in the gray histogram and the number of pixel points of the gray level in the template image When the obtained absolute value is not 0, the gray level is considered to be a differential gray level, otherwise the gray level is not a differential gray level. The various gray levels of each gray image are sequentially judged to obtain the various differential gray levels of each gray image. The various differential gray levels in each gray image and the number of pixel points corresponding to the various differential gray levels are recorded, and the set composed of all the differential gray levels in the s-th gray image is denoted as , wherein the i-th differential gray level is denoted as , The number of pixel points corresponding to the i-th differential gray level is denoted as The light quality evaluation index of the s-th gray image can be represented as:

[0038]

[0039] In the formula, represents the number of pixel points corresponding to the i-th differential gray level of the s-th gray image; represents the gray value of the i-th differential gray level of the s-th gray image; represents the total number of differential gray levels contained in the set ; and is the total number of pixel points corresponding to all the differential gray levels contained in the set ; and is a hyperbolic tangent function; is a logarithmic function with base 2.

[0040] ​​​The average gray value representing all differential gray levels of the s-th gray image and the template image, since the scene collected by the monitoring camera is fixed, the collection angle of the monitoring camera is fixed, and the position of the light source is fixed, only when there is a moving object in the monitoring picture, the gray value of part of the pixel points in the image will change, at this time, the pixel points with changed gray value are the pixel points corresponding to the moving object. Since the average gray value represents the gray difference between the moving area caused by the movement of the medical staff in the s-th gray image and the corresponding area of the template image, since the scene is fixed when the monitoring camera collects the image, the smaller the gray difference between the moving area formed by the medical staff and the corresponding area of the template image, the weaker the light in the moving area formed by the medical staff, which makes the gray difference between the moving area and the corresponding area in the template image smaller, that is, the brightness of the area where the medical staff is located in the s-th gray image is lower, at this time, the worse the light quality, the smaller the light quality evaluation index of the s-th gray image, and vice versa.

[0041] The ratio between the number of pixel points corresponding to the i-th differential gray level of the s-th gray image and the total number of pixel points corresponding to all differential gray levels in the s-th gray image, which represents the frequency of the i-th differential gray level; The ratio between the number of pixel points corresponding to the i-th differential gray level of the s-th gray image and the total number of pixel points corresponding to all differential gray levels in the s-th gray image, which represents the frequency of the i-th differential gray level; The information entropy representing the moving area corresponding to the medical staff in the s-th gray image, which is used to represent the structural richness of the area corresponding to the medical staff in the s-th gray image. The more the types of gray levels, the higher the gray richness, and the more detailed information contained in the area corresponding to the medical staff. When the medical staff moves from a brighter area to a darker area, the detailed information in the area corresponding to the medical staff is lost due to the decrease in light quality, which leads to a significant decrease in the gray richness in the area. Therefore, the average gray value of all differential gray levels of the s-th gray image and the structural richness of the area corresponding to the medical staff are used to describe the light quality of each monitoring image during the movement of the medical staff. Then, the average gray value of all differential gray levels of the s-th gray image and the structural richness of the area corresponding to the medical staff are normalized using the hyperbolic tangent function, so that the values of both are limited to 0-1. Finally, the sum is taken to obtain the light quality evaluation index of the s-th image. At this time, the smaller the average gray value of all differential gray levels of the s-th gray image, the smaller the structural richness, which indicates that the light quality of the area corresponding to the medical staff in the s-th gray image is worse, and the smaller the light quality evaluation index obtained at this time.

[0042] The above method is repeated to obtain the light quality evaluation index of each gray image.

[0043] The sensitive factor acquisition module is used to acquire each pixel point in each gray image in which information loss occurs, so as to obtain the sensitive factor in each gray image.

[0044] The embodiment aims to compensate the gray of the image in which detail loss has occurred, so as to restore the lost detail information, so as to ensure the integrity of the image information, thus the embodiment first needs to judge which gray levels are more likely to disappear when the light quality changes, and these gray levels which are more likely to disappear are the sensitive factors which need to be compensated in the embodiment. Since the light quality evaluation index is acquired according to the pixel point in which the gray changes, and the reason for the gray change is the movement of the medical staff, thus the light quality evaluation index represents the light quality of the position where the medical staff is located, when the light quality gradually increases, the detail information of the corresponding region of the medical staff will also gradually increase, and the types of gray levels will also gradually increase, at this time, the detail information of the medical staff is also gradually increased, and it does not need to be compensated for the gray; when the light quality decreases, since the gray image corresponding to the last moment at this moment contains more detail information, thus the detail information of the corresponding position can be extracted from the adjacent gray image corresponding to the last moment, and the position in which the detail loss occurs in the gray image at this moment is compensated for the gray, so as to improve the image quality.

[0045] The movement process of the medical staff in the monitoring video is continuous movement, thus the embodiment first acquires the target frame in each gray image by using the target tracking algorithm, and acquires each gray level in the target frame corresponding to each gray image, wherein the gray level set composed of each gray level in the target frame corresponding to the s-th gray image is denoted as In order to reduce the calculation amount, the subsequent process only needs to analyze the pixel points in the target frame in each gray image.

[0046] Then the gray levels which disappear in the gray image corresponding to the current s-th moment relative to the gray image corresponding to the last moment, i.e. the s-1-th moment are determined, that is, the sensitive factor in the gray image corresponding to the current s-th moment is determined, and the difference between the current moment and the last moment lies in the difference of the light quality, thus the embodiment restores the light quality of the current moment to the illumination quality of the last moment according to the relative size between the light quality of the current moment and the light quality of the last moment, so as to obtain the ideal gray level corresponding to each gray level of the current s-th moment under the ideal condition if the light quality does not change; at this time, only the difference between each ideal gray level and each gray level of the s-1-th moment is needed, so as to find the gray levels which disappear in the gray image corresponding to the s-th moment relative to the gray image of the last moment, so as to find the sensitive factor of the gray image corresponding to the s-th moment, and the gray image corresponding to the s-th moment is the s-th gray image in the above.

[0047] the degree of change of the light quality of the current s-th gray-scale image relative to the s-1-th gray-scale image may be expressed as:

[0048]

[0049] In the formula, is the light quality evaluation index of the s-th gray-scale image, is the light quality evaluation index of the s-1-th gray-scale image.

[0050] When is less than 1, it indicates that the light quality at the current s-th moment is reduced relative to the light quality at the previous moment, which indicates that the detail information in the corresponding region of the medical staff at the current moment is gradually reduced relative to the previous moment, and at this time, the sensitive factor in the corresponding gray-scale image at the s-th moment needs to be compensated for gray scale; when is greater than or equal to 1, it indicates that the detail information in the corresponding region of the medical staff at the current moment is gradually increased or basically unchanged relative to the previous moment, and at this time, the corresponding gray-scale image at the s-th moment does not need to be compensated for gray scale.

[0051] When is less than 1, the sensitive factor in the s-th gray-scale image needs to be compensated for gray scale, and the specific process is as follows:

[0052] First, multiply the value of each gray level inside the target selection box region in the s-th gray-scale image corresponding to the current s-th moment, i.e., the s-th gray-scale image, by and then take the integer part, to obtain each gray level after the light quality of the s-th gray-scale image is restored to the light quality of the previous frame image. These gray levels are the ideal gray levels of the s-th gray-scale image, and each ideal gray level of the s-th gray-scale image constitutes an ideal gray level set of the s-th gray-scale image when the light quality is the same as that of the s-1-th gray-scale image At this time, by comparing with the gray level set in the target selection box in the s-1-th frame image This allows us to obtain the gray levels that disappear due to changes in light in the (s-1)th grayscale image. In adjacent frames of the surveillance video, the movement of medical staff is stable, and their movement speed is much slower than the acquisition speed of the surveillance camera. This means that the positions of medical staff in adjacent grayscale images have not changed significantly. The target selection boxes of medical staff in adjacent grayscale images have similar shapes and sizes. Therefore, in this embodiment, we consider all pixels corresponding to the gray levels that disappeared in the target selection box of the (s-1)th grayscale image, and the pixels with the same coordinates as the sensitive factors in the target selection box of the sth grayscale image, to be the sensitive factors in the sth grayscale image. Subsequently, grayscale compensation is performed on these sensitive factors to ensure image quality.

[0053] To avoid the influence of error, calculation Each gray level and The difference between the various gray levels in the image, for example, for For the t-th gray level in the image, obtain the gray level and its relationship with the image. The minimum value of the difference between each gray level in the image. ,when If the gray level does not disappear due to changes in light quality, then the pixel corresponding to that gray level is not a sensitive factor. Otherwise, the pixel corresponding to that gray level is considered a sensitive factor of the (s-1)th gray-level image, and this process is repeated sequentially. From each gray level in the image, we obtain all the sensitivity factors in the s-th grayscale image.

[0054] The grayscale compensation module is used to compensate for the sensitivity factors in each grayscale image, thereby improving the image quality of each grayscale image.

[0055] The gray values ​​of each pixel in the 8-neighborhood of each sensitive factor corresponding to the s-th gray image are called the structural information of each sensitive factor. That is, with each sensitive factor as the center point of the sliding window, the gray values ​​of all pixels in the 3×3 sliding window of each sensitive factor are obtained, and the gray values ​​of these pixels are called the structural information of each sensitive factor.

[0056] Then, obtain the structural information of the corresponding pixel in the target selection box of the (s-1)th grayscale image, and calculate the structural similarity between the structural information of each sensitivity factor in the s-th grayscale image and the structural information of the corresponding pixel in the (s-1)th grayscale image. The structural similarity is a well-known technique and will not be elaborated here.

[0057] Based on experience, a threshold is set; in this embodiment, the threshold is 0.8. When the structural similarity between a sensitive factor and its corresponding pixel is greater than or equal to the threshold, grayscale compensation is performed on the sensitive factor. The grayscale value of the r-th sensitive factor corresponding to the s-th grayscale image after grayscale compensation is denoted as... ,but:

[0058]

[0059] In the formula, represents the rth sensitive factor gray value of the st gray scale image, represents the gray value of the corresponding pixel point of the rth sensitive factor in the s-1th gray scale image, is a hyperparameter, which is used to represent the reference weight of the corresponding pixel point in the adjacent gray scale image, and in the embodiment .

[0060] The above method is repeated to compensate the sensitive factors of each gray scale image, and the gray compensated image is used for the calculation and gray compensation of the sensitive factors of the next gray scale image until the medical staff disappears from the screen.

[0061] The risk detection module is used to detect the missing risks of each gray scale image after gray compensation.

[0062] Each gray scale image after gray compensation is input into the risk detection module, that is, the identification of the missed small waste, the face recognition of the non-related personnel approaching, and the identification of the dangerous operation such as the medical staff running and the waste storage barrel being dropped. Since the integrity of the image detail information is ensured in each gray scale image after gray compensation, the problem of detail loss caused by light change can be improved, so that more reliable analysis results can be obtained when monitoring and analyzing the risks according to the monitoring video image, and the accuracy of the missing risk monitoring is improved.

[0063] Through the above steps, the missing risks of the medical waste cold storage are detected.

[0064] The embodiment first calculates the light quality of different images of the monitoring video image, that is, the light quality evaluation index; the light quality of each gray image is restored to the corresponding light quality at the previous time through the light quality evaluation index of each gray image and its adjacent gray image, the ideal gray level corresponding to each gray level in the target frame of each gray image under the condition of not being disturbed by the change of light quality is obtained, and further according to the difference between the ideal gray level and each gray level in the target frame corresponding to the actual previous time, the gray level disappeared in the previous time due to the change of light quality is obtained, so as to obtain the sensitive factor in each gray image which is easy to be lost due to the change of light. Finally, the gray value of the sensitive factor of each gray image is weighted with the gray value of the pixel point at the corresponding position in the previous time, so as to realize the gray compensation of the dynamic target image in the uneven illumination scene, restore the lost detail information in the image, ensure the integrity of the image detail information, and improve the image quality. Compared with the image preprocessing method of the traditional monitoring system, the method can greatly restore the easily lost detail information in the image, and compared with the similar compensation algorithm such as interpolation, the compensation effect obtained by the method is more accurate, the integrity and reliability of the monitoring video information are improved, and the result obtained by the subsequent security, omission, illegal operation and other risk detection of the medical waste transportation process is more accurate.

[0065] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A system for monitoring the risk of omissions in a medical waste cold store, characterized in that The system comprises the following modules: A video data acquisition module acquires monitoring video of medical staff transporting medical waste to obtain continuous gray images, and takes the gray images when no one is present as a template image; A light quality evaluation module obtains differential gray levels of each gray image according to a gray histogram of each gray image and a gray histogram of the template image; and obtains a light quality evaluation index of each gray image according to an average gray value of each differential gray level in each gray image and structural richness in each gray image; A sensitive factor acquisition module acquires a target frame in each gray image; and obtains a sensitive factor of each gray image according to the light quality evaluation index of each gray image and a neighboring gray image; wherein, each pixel point corresponding to each vanishing gray level in the neighboring gray image at a corresponding position in each gray image is recorded as each sensitive factor of each gray image; A gray compensation module compensates the gray value of each sensitive factor according to the gray value of each sensitive factor of each gray image and the gray value of a pixel point at a corresponding position in a neighboring gray image of each gray image, to obtain a gray value of each sensitive factor after gray compensation; and sequentially compensates the gray value of each gray image; A risk detection module detects each gray image after gray compensation for missing risk; The light quality evaluation index is expressed as: ； In the formula, represents the light quality evaluation index of the s-th gray image, represents the number of pixel points corresponding to the i-th differential gray level of the s-th gray image; represents the gray value of the i-th differential gray level of the s-th gray image; represents a set composed of all differential gray levels in the s-th gray image; represents the total number of differential gray levels contained in the set; is the total number of pixel points corresponding to all differential gray levels contained in the set; is a hyperbolic tangent function; is a logarithmic function with base 2; an average gray value representing all differential gray levels of the s-th gray image and the template image; a ratio representing the number of pixel points corresponding to the i-th differential gray level in the s-th gray image and the total number of pixel points in the s-th gray image, indicating the frequency of the i-th differential gray level in the s-th gray image; a ratio between the total number of pixel points corresponding to all differential gray levels in the s-th gray image and the total number of pixel points in the s-th gray image, indicating the frequency of the i-th differential gray level in the s-th gray image; an information entropy representing the corresponding motion area of the medical staff in the s-th gray image, used to represent the structural richness of the corresponding area of the medical staff in the s-th gray image; The method for obtaining the differential gray level of each gray image is as follows: The number of pixel points corresponding to each gray level in the gray histogram of each gray image and the gray histogram of the template image is subtracted, and the absolute value of the obtained difference value is taken; when the absolute value is not 0, the gray level is a differential gray level, otherwise the gray level is not a differential gray level; The method for obtaining the sensitive factor of each gray image according to the light quality evaluation index of each gray image and a neighboring gray image comprises: An ideal gray level set in a target frame of each gray image is obtained according to the light quality evaluation index of each gray image and a neighboring gray image; and a sensitive factor of each gray image is obtained according to the ideal gray level set corresponding to each gray image and a gray level set in the target frame of a neighboring gray image; The method for obtaining the ideal gray level set in the target frame of each gray image is as follows: A light quality change degree of each gray image is obtained by taking the ratio between the light quality evaluation index of a neighboring gray image corresponding to each gray image and the light quality evaluation index of each gray image; each ideal gray level of each gray image is obtained by multiplying the light quality change degree and each gray level contained in the target frame of each gray image, and then taking the result down to the nearest integer; and the ideal gray levels constitute an ideal gray level set of each gray image; The vanishing gray level is The absolute value of the difference between each ideal gray level corresponding to each gray image and each gray level in the adjacent gray image of the each gray image is calculated, and when the absolute value is less than or equal to 5, the each ideal gray level is a disappearing gray level of the adjacent gray image; otherwise, the each ideal gray level is not a disappearing gray level of the adjacent gray image.

2. The system for monitoring the risk of omissions for medical waste cold stores of claim 1, characterized in that, The method for obtaining the gray value of each sensitive factor after compensation is: An over-parameter is set, and the difference between 1.0 and the over-parameter is used as a reference weight of each sensitive factor gray value of each gray image; The over-parameter is used as a reference weight of the corresponding pixel point gray value of each sensitive factor in the adjacent gray image of each gray image; The gray value of each sensitive factor and the gray value of the corresponding pixel point are weighted and summed, and the obtained result is used as the gray value of each sensitive factor after compensation.

Images