Remnant rapid detection algorithm and system based on smart park machine room environment

By designing a rapid legacy detection algorithm based on difference comparison and clustering processing in the computer room environment of smart parks, the problems of complexity and resource consumption of legacy detection in complex environments are solved, and automated and accurate legacy recognition and adaptation are achieved. Different environmental requirements.

CN118918526AInactive Publication Date: 2024-11-08SHANDONG MODERN BIG DATA TECH CO LTD

Patent Information

Application Number
CN202410708834.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-03
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In a complex and crowded environment, the existing legacy detection algorithm is complex, resulting in large resource consumption and extended discovery time, making it difficult to quickly identify legacy in the computer room, affecting the efficiency of the owner to retrieve items.

Method used

A rapid detection algorithm for legacy is designed based on the computer room environment of the smart park. By selecting the difference between the background image and the detected image, the outline data is extracted, the outline data is clustered and merged, and the threshold parameters and automated adjustments are combined to quickly identify legacy. And through image acquisition, background modeling, contour analysis and parameter optimization modules working together to achieve automated detection.

Benefits of technology

It realizes automated inspection, reduces the workload of manual inspection, improves detection accuracy, saves resource costs, adapts to different environmental needs, and quickly identifies the remains in the computer room.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918526A_ABST
    Figure CN118918526A_ABST
Patent Text Reader

Abstract

The invention relates to the field of detection algorithms, and discloses a remnant rapid detection algorithm and system based on a smart park machine room environment, and the method comprises the steps: taking a picture captured from a camera video stream as a detected picture, carrying out the difference comparison of the detected picture and a background picture, and obtaining a difference picture; performing transformation processing on the difference picture, and extracting the contour of the difference part; and performing clustering operation on the contours according to the distance, combining a plurality of object contours in the same type, selecting the outmost contour of a combination result as a final remnant identification contour, and drawing the outmost contour on the detected picture as a final detection result. According to the remnant rapid detection algorithm and system based on the smart park machine room environment, by optimizing the algorithm and introducing clustering, merging and other technologies, misjudgment caused by irregular objects can be reduced, the accuracy of object recognition is improved, the functions of automatically adjusting parameters and optimizing the algorithm are achieved, and the detection efficiency is improved. Intelligent management can be carried out according to different environment characteristics, and the requirements of different machine room environments are met.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of detection algorithms, and specifically to a rapid detection algorithm and system for relics based on a smart park computer room environment. Background Art

[0002] The server room is where servers are stored and run. It is the core of modern information systems and the key to whether information systems can operate normally. Therefore, ensuring the cleanliness of the room environment is one of the important tasks of daily inspections of the server room.

[0003] With the development of information technology, especially the development of machine vision technology and deep learning technology, object recognition based on camera video is becoming more and more widely used in all walks of life. On the one hand, this technology has high accuracy and timeliness, and on the other hand, it greatly reduces the workload of relevant personnel and improves work efficiency.

[0004] Object detection is one aspect of machine vision object recognition applications, especially in crowded areas such as airports, stations, and shopping malls. By detecting objects, lost items can be found as soon as possible, helping the owner to find the lost items, thereby reducing losses. Therefore, there are many object detection methods and applications based on various technologies on the market. However, areas such as airports, stations, and shopping malls have complex environments and dense crowds, which puts high demands on object detection. Therefore, object detection algorithms for such environments are often more complex. Not only do they need to track the positional relationship between people and objects, but they also need to leave enough time for the objects to exist alone to ensure that the objects are indeed lost, rather than deliberately stored temporarily. These complex processes increase resource consumption, prolong the time it takes to find objects, and make it more difficult for the owner to find their lost items.

[0005] Compared with airports, train stations, and shopping malls, the server room is a relatively simple environment. There is no complicated flow of people and vehicles. Generally, there will be designated working hours and staff. Outside of working hours, the room environment must be kept clean and tidy, and other people are not allowed to enter at this time. Therefore, for this relatively simple environment, a simpler legacy detection algorithm can be used to reduce resource consumption and increase detection speed. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides a rapid detection algorithm and system for relics based on a smart park computer room environment to solve the above problems.

[0008] (II) Technical solution

[0009] To achieve the above object, the present invention provides the following technical solution: a rapid detection algorithm for residues based on the environment of a smart park computer room, comprising the following steps:

[0010] S1. Select a background image: select a picture without people and leftover items as the background image, the background image is clear and unobstructed;

[0011] S2. Capture the detected image: Capture an image from the camera video stream as the detected image. The detected image has high quality and moderate resolution, and can accurately capture changes in the machine room;

[0012] S3. Compare difference pictures: compare the detected picture and the background picture to obtain a difference picture;

[0013] S4. Extracting contour data: transforming the difference image to extract the contour of the difference part;

[0014] S5. Clustering and merging contours: clustering contours based on distance and merging contours of multiple objects in the same category;

[0015] S6. Draw the outline of the remnant: select the outermost outline of the merged result as the final remnant recognition outline and draw it on the detected image;

[0016] S7. Send test results: send the test results to the business system.

[0017] Preferably, a threshold parameter is introduced in S4, and can be adjusted and optimized according to different environments to obtain optimal contour data.

[0018] Preferably, the specific process of introducing the threshold parameter is as follows:

[0019] Determine the threshold parameters: Select the threshold parameters according to the specific environment's lighting, background interference, etc., set the initial threshold based on experience, and adjust it based on the actual effect.

[0020] Threshold processing: During the image processing, the selected threshold parameters are used to process the difference pictures, and the threshold segmentation technology is used to convert the image into a binary image to highlight the difference between the remains and the background.

[0021] Contour extraction: On the image after threshold processing, the contour data of the remains is extracted, and the edge detection algorithm is used to obtain clear contour information.

[0022] Parameter optimization: Evaluate the quality of contour data based on actual results, adjust threshold parameters based on the evaluation results, and re-extract contour data until the optimal contour data is obtained.

[0023] Automatic adjustment: Introduce an automatic adjustment mechanism to automatically adjust threshold parameters through algorithms, achieve optimization based on feedback results, and adopt an adaptive threshold processing method to dynamically adjust parameters to adapt to different environments.

[0024] Preferably, a clustering and merging process is introduced in S5 contour analysis to reduce misjudgment caused by irregular objects and improve the accuracy of object recognition.

[0025] Preferably, the specific process of introducing the clustering and merging process is as follows:

[0026] Contour clustering: clustering the extracted contour data to classify contours of similar shapes and sizes into the same category;

[0027] Merge similar contours: In each clustering result, merge adjacent contours to eliminate misjudgments caused by irregular shapes. Merge operations are performed based on the distance and shape similarity between contours.

[0028] Select representative contours: Select representative contours in each cluster as the final recognition result of the category, and select the contour with the largest contour area or other obvious features as the representative;

[0029] Optimize parameter adjustment: Evaluate clustering and merging results based on actual results, adjust clustering parameters and merging strategies to obtain the best object recognition effect, and optimize parameters based on indicators such as error rate and accuracy.

[0030] Preferably, the entire algorithm has no correlation with the previous and subsequent time, and can make judgments immediately and quickly identify objects that are not in the background.

[0031] According to another aspect of an embodiment of the present invention, a rapid detection system for relics based on a smart campus computer room environment is provided, including: an image acquisition module, a background modeling module, a relics detection module, a contour analysis module, a parameter optimization module, a result display module and a log recording module. Through the collaborative work between the modules, rapid and accurate detection of relics in the computer room can be achieved.

[0032] Preferably, the image acquisition module is used to obtain real-time images or video streams from the camera, the background modeling module is used to model the background image of the computer room for use as a reference for relics detection, the relics detection module is used to detect relics in the computer room based on the difference between the background image and the real-time image, the contour analysis module is used to cluster, merge, and other processes the detected relics contours to improve detection accuracy, the parameter optimization module is used to automatically adjust threshold parameters and clustering parameters according to different environmental characteristics to optimize the detection algorithm effect, the result display module is used to display the detection results in real time on the interface for user observation and analysis, and the log recording module is used to record system operation logs, including detection results and parameter adjustment records, for system performance analysis and troubleshooting.

[0033] (III) Beneficial effects

[0034] Compared with the prior art, the present invention provides a rapid detection algorithm and system for residues based on the environment of a smart park computer room, which has the following beneficial effects:

[0035] This rapid detection algorithm and system for relics based on the environment of the computer room of a smart park can automatically detect relics, reduce the workload of manual inspections, and save human resource costs. By optimizing algorithms and introducing clustering, merging and other technologies, it can reduce misjudgments caused by irregular objects and improve the accuracy of object recognition. It has the function of automatically adjusting parameters and optimizing algorithms, and can perform intelligent management according to the characteristics of different environments to adapt to the needs of different computer room environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 The figure is a schematic diagram of the overall workflow of the rapid detection algorithm and system of relics based on the smart park computer room environment of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0039] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0040] See also Figure 1 ,The rapid detection algorithm of the remains based on the ,smart park computer room environment and ,includes the following steps:

[0041] S1. Select a background image: Select a picture without people and objects left behind as the background image. The background image should be clear and unobstructed.

[0042] S2. Capture the detected image: Capture an image from the camera video stream as the detected image. The detected image has high quality and moderate resolution, and can accurately capture the changes in the computer room.

[0043] S3. Compare difference pictures: compare the detected picture and the background picture to obtain a difference picture;

[0044] S4. Extracting contour data: transforming the difference image to extract the contour of the difference part;

[0045] S5. Clustering and merging contours: clustering contours based on distance and merging contours of multiple objects in the same category;

[0046] S6. Draw the outline of the remnant: select the outermost outline of the merged result as the final remnant recognition outline and draw it on the detected image;

[0047] S7. Send test results: send the test results to the business system.

[0048] Among them, the threshold parameter is introduced in S4, which can be adjusted and optimized according to different environments to obtain the optimal contour data.

[0049] The specific process of introducing the threshold parameter is as follows:

[0050] Determine the threshold parameters: Select the threshold parameters according to the specific environment's lighting, background interference, etc., set the initial threshold based on experience, and adjust it based on the actual effect.

[0051] Threshold processing: During the image processing, the selected threshold parameters are used to process the difference pictures, and the threshold segmentation technology is used to convert the image into a binary image to highlight the difference between the remains and the background.

[0052] Contour extraction: On the image after threshold processing, the contour data of the remains is extracted, and the edge detection algorithm is used to obtain clear contour information.

[0053] Parameter optimization: Evaluate the quality of contour data based on actual results, adjust threshold parameters based on the evaluation results, and re-extract contour data until the optimal contour data is obtained.

[0054] Automatic adjustment: Introduce an automatic adjustment mechanism to automatically adjust threshold parameters through algorithms, achieve optimization based on feedback results, and adopt an adaptive threshold processing method to dynamically adjust parameters to adapt to different environments.

[0055] Among them, clustering and merging processes are introduced in S5 contour analysis to reduce misjudgment caused by irregular objects and improve the accuracy of object recognition.

[0056] Among them, the specific process of introducing clustering and merging process is as follows:

[0057] Contour clustering: clustering the extracted contour data to classify contours of similar shapes and sizes into the same category;

[0058] Merge similar contours: In each clustering result, merge adjacent contours to eliminate misjudgments caused by irregular shapes. Merge operations are performed based on the distance and shape similarity between contours.

[0059] Select representative contours: Select representative contours in each cluster as the final recognition result of the category, and select the contour with the largest contour area or other obvious features as the representative;

[0060] Optimize parameter adjustment: Evaluate clustering and merging results based on actual results, adjust clustering parameters and merging strategies to obtain the best object recognition effect, and optimize parameters based on indicators such as error rate and accuracy.

[0061] The entire algorithm has no correlation with the previous and subsequent time, and can make immediate judgments and quickly identify objects that are not in the background.

[0062] According to another aspect of an embodiment of the present invention, a rapid detection system for relics based on a smart campus computer room environment is provided, including: an image acquisition module, a background modeling module, a relics detection module, a contour analysis module, a parameter optimization module, a result display module and a log recording module. Through the collaborative work between the modules, rapid and accurate detection of relics in the computer room can be achieved.

[0063] Among them, the image acquisition module is used to obtain real-time images or video streams from the camera, the background modeling module is used to model the background image of the computer room for use as a reference for relics detection, the relics detection module is used to detect relics in the computer room based on the differences between the background image and the real-time image, the contour analysis module is used to cluster, merge and perform other processing on the detected relics contours to improve the detection accuracy, the parameter optimization module is used to automatically adjust the threshold parameters and clustering parameters according to the characteristics of different environments to optimize the detection algorithm effect, the result display module is used to display the detection results in real time on the interface for user observation and analysis, and the log recording module is used to record the system operation log, including the detection results and parameter adjustment records, which are used for system performance analysis and troubleshooting.

[0064] In summary, the present invention provides a rapid detection algorithm and system for relics based on the smart park computer room environment, which can automatically perform relics detection, reduce the workload of manual inspections, and save human resource costs. By optimizing algorithms and introducing clustering, merging and other technologies, it can reduce misjudgments caused by irregular objects and improve the accuracy of object recognition. It has the function of automatically adjusting parameters and optimizing algorithms, and can perform intelligent management according to the characteristics of different environments to adapt to the needs of different computer room environments.

[0065] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Rapid detection algorithm of relics based on the environment of smart park computer room, It is characterized in that The following steps are involved: S1. Select a background image: select a picture without people and leftover items as the background image, the background image is clear and unobstructed; S2. Capture the detected image: Capture an image from the camera video stream as the detected image. The detected image has high quality and moderate resolution, and can accurately capture changes in the machine room; S3. Compare difference pictures: compare the detected picture and the background picture to obtain a difference picture; S4. Extracting contour data: transforming the difference image to extract the contour of the difference part; S5. Clustering and merging contours: clustering contours based on distance and merging contours of multiple objects in the same category; S6. Draw the outline of the remnant: select the outermost outline of the merged result as the final remnant recognition outline and draw it on the detected image; S7. Send test results: send the test results to the business system.

2. According to claim 1, the rapid detection algorithm for relics based on the environment of the smart park computer room, Features: The threshold parameters introduced in S4 can be adjusted and optimized according to different environments to obtain the optimal contour data.

3. According to claim 2, the rapid detection algorithm for relics based on the environment of the smart park computer room, Features: The specific process of introducing threshold parameters is as follows: Determine the threshold parameters: Select the threshold parameters according to the specific environment's lighting, background interference, etc., set the initial threshold based on experience, and adjust it based on the actual effect. Threshold processing: During the image processing, the selected threshold parameters are used to process the difference pictures, and the threshold segmentation technology is used to convert the image into a binary image to highlight the difference between the remains and the background. Contour extraction: On the image after threshold processing, the contour data of the remains is extracted, and the edge detection algorithm is used to obtain clear contour information. Parameter optimization: Evaluate the quality of contour data based on actual results, adjust threshold parameters based on the evaluation results, and re-extract contour data until the optimal contour data is obtained. Automatic adjustment: Introduce an automatic adjustment mechanism to automatically adjust threshold parameters through algorithms, achieve optimization based on feedback results, and adopt an adaptive threshold processing method to dynamically adjust parameters to adapt to different environments.

4. According to claim 1, the rapid detection algorithm for relics based on the environment of the smart park computer room, Features: Clustering and merging processes are introduced in S5 contour analysis to reduce misjudgments caused by irregular objects and improve the accuracy of object recognition.

5. According to claim 5, the rapid detection algorithm for relics based on the environment of the smart park computer room, Features: The specific process of introducing clustering and merging process is as follows: Contour clustering: clustering the extracted contour data to classify contours of similar shapes and sizes into the same category; Merge similar contours: In each clustering result, merge adjacent contours to eliminate misjudgments caused by irregular shapes. Merge operations are performed based on the distance and shape similarity between contours. Select representative contours: Select representative contours in each cluster as the final recognition result of the category, and select the contour with the largest contour area or other obvious features as the representative; Optimize parameter adjustment: Evaluate clustering and merging results based on actual results, adjust clustering parameters and merging strategies to obtain the best object recognition effect, and optimize parameters based on indicators such as error rate and accuracy.

6. According to claim 1, the rapid detection algorithm for relics based on the environment of the smart park computer room, Features: The entire algorithm has no correlation with the previous and subsequent time, and can make immediate judgments and quickly identify objects that are not in the background.

7. Rapid detection system for leftovers based on the environment of the smart park computer room, include: The image acquisition module, background modeling module, residue detection module, contour analysis module, parameter optimization module, result display module and log recording module are characterized in that the rapid and accurate detection of residues in the computer room can be achieved through the collaborative work among the modules.

8. According to claim 7, the rapid detection system for residues based on the environment of the smart park computer room, Features: The image acquisition module is used to obtain real-time images or video streams from the camera, the background modeling module is used to model the background image of the computer room for use as a reference for relics detection, the relics detection module is used to detect relics in the computer room based on the difference between the background image and the real-time image, the contour analysis module is used to cluster, merge and perform other processing on the detected relics contours to improve detection accuracy, the parameter optimization module is used to automatically adjust threshold parameters and clustering parameters according to different environmental characteristics to optimize the detection algorithm effect, the result display module is used to display the detection results in real time on the interface for user observation and analysis, and the log recording module is used to record system operation logs, including detection results and parameter adjustment records, for system performance analysis and troubleshooting.

Citation Information

Patent Citations

  • A plot extraction method and an apparatus for a remote sensing image

    CN109409176A

  • Quad-rotor unmanned aerial vehicle autonomous landing method based on visual positioning

    CN110569838A

  • Clustering method, system and device for moving target detection in video and medium

    CN117496399A

  • Automatic driving target detection method based on incremental small sample learning

    CN117612136A

  • Method for detecting articles left in carry-on luggage basket of passenger

    CN118033768A

Cited By

  • Anti-misjudgment image analysis operation and maintenance system

    CN120107534A