Intelligent automatic evaluation system and method for intelligent prison

Abstract

The invention relates to an intelligent automatic evaluation system and method for an intelligent prison, and the system comprises a daily evaluation module, an evaluation record affirmation evaluation module and an evaluation information storage module which are connected in sequence. The daily evaluation module is used for evaluating the monitored abnormal behaviors found by AI calculation and manual inspection to form a final daily evaluation result; the evaluation record affirmation and evaluation module is used for affirmation, evaluation and scoring according to the daily evaluation results, and the evaluation information storage module is used for storing the daily evaluation results and performing final evaluation summarization. According to the method, the evaluation result is automatically calculated in a manual or AI labeling mode, so that the evaluation efficiency is greatly improved. Compared with a traditional manual evaluation calculation mode, the AI evaluation system can process more data and information within the same time, and therefore the large-scale and high-efficiency evaluation requirement is met.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent evaluation technology, and in particular to an intelligent automatic evaluation system and method for smart prisons. Background Art

[0002] In the detention center, it is an important task for police officers to evaluate the performance of detainees, which aims to ensure the stability of the detention order, promote the active transformation of detainees, and prepare them for reintegration into society. In their daily work, police officers record and evaluate the behavior of detainees and form daily evaluation records. Detention centers can set a certain evaluation cycle (such as monthly, quarterly, etc.) to conduct regular evaluations of detainees and comprehensively evaluate their performance during detention.

[0003] Defects and shortcomings of the existing technology:

[0004] When the police are short of manpower, the assessment and evaluation of the supervised persons will not be timely, which will not only affect the standardization of the police work, but also lead to negligence in the assessment and management of the supervised persons;

[0005] After the police have registered for the evaluation, they usually need the approval of their leaders. During the approval process, multiple parties (supervised persons, supervising police officers, and station leaders) need to sign and stamp. The paper report format wastes resources and manpower.

[0006] After the police evaluate the detention performance of the supervised person, if you want all the file information, you have to sort and summarize it before you can see all the evaluation records. Summary of the invention

[0007] The purpose of the present invention is to overcome the shortcomings of the prior art and propose an intelligent automatic evaluation system and method for smart prisons, which automatically calculates the evaluation results through manual or AI labeling, greatly improving the evaluation efficiency. Compared with the traditional manual evaluation calculation method, the AI ​​evaluation system can process more data and information in the same time, thereby meeting the needs of large-scale and efficient evaluation.

[0008] The present invention solves the technical problem by adopting the following technical solutions:

[0009] The intelligent automatic evaluation system for smart prisons includes a daily evaluation module, an evaluation record recognition and evaluation module and an evaluation information storage module, wherein the daily evaluation module, the evaluation record recognition and evaluation module and the evaluation information storage module are connected in sequence. The daily evaluation module is used to evaluate the abnormal behaviors of the supervised persons discovered by AI calculations and manual inspections to form the final daily evaluation results; the evaluation record recognition and evaluation module is used to perform recognition and evaluation scoring according to the daily evaluation results; and the evaluation information storage module is used to store the results of daily evaluations and make final evaluation summaries.

[0010] An evaluation method for an intelligent automatic evaluation system for a smart prison comprises the following steps:

[0011] Step 1: The daily evaluation module evaluates the abnormal behaviors of supervised personnel found by AI calculations and manual inspections to form the final daily evaluation results;

[0012] Step 2: The evaluation record recognition evaluation module is used to perform recognition evaluation scoring based on the daily evaluation results;

[0013] Step 3: The evaluation information storage module is used to store the results of daily evaluations and to make final evaluation summaries.

[0014] Moreover, the specific implementation method of step 1 is as follows: using AI algorithm (analyzing the video in the cell, detecting abnormal behavior of the supervised person and pushing it to the alarm management service, which pushes it to the supervision super brain service after receiving it, and the supervision super brain service records it in the daily evaluation of the supervised person according to the pre-set rules; at the same time, the patrol post manually marks the abnormal behavior of the supervised person during the online patrol, and the generated abnormal behavior label of the supervised person will also be recorded in the daily evaluation. The police at the disciplinary post evaluate the abnormal behavior of the supervised person found by AI calculation and manual patrol according to the actual situation, form the final daily evaluation result, and push it to the leader for subsequent approval;

[0015] Leaders review and approve daily evaluation records pushed by the disciplinary post, and those approved are confirmed by the supervised persons on the in-cell screen; if the supervised persons do not apply for reconsideration, the daily evaluation information will be recorded in the supervised persons' files; if the supervised persons apply for reconsideration, the leader needs to conduct reconsideration and approval, and after approval, the daily evaluation information will be recorded in the supervised persons' files, and daily evaluation-related documents will be generated simultaneously.

[0016] Moreover, the specific implementation method of step 2 is: the disciplinary post screens the daily evaluation records and identification evaluation categories of the supervised persons for a period of time, generates the identification evaluation records, the leader of the station conducts the final review, adds approval opinions, generates the identification evaluation of the supervised persons, and simultaneously generates an electronic evaluation document; at the same time, the evaluation content is formulated, the supervised persons check the formula table on the screen in the prison, the disciplinary police officers query the supervised persons' review records, and the identification evaluation and evaluation formulas are used to simultaneously generate evaluation-related documents.

[0017] Moreover, the specific implementation method of step 3 is: the police can choose to create a detention performance file for the supervised person. After the file is created, it will automatically collect the basic information of the supervised person, daily evaluation records, identification evaluation records, evaluation formula records and corresponding relevant documents in the records.

[0018] The advantages and positive effects of the present invention are:

[0019] 1. The present invention includes a daily evaluation module, an evaluation record identification evaluation module and an evaluation information storage module, wherein the daily evaluation module, the evaluation record identification evaluation module and the evaluation information storage module are connected in sequence, and the daily evaluation module is used to evaluate the abnormal behaviors of supervision discovered by AI calculation and manual inspection to form the final daily evaluation results; the evaluation record identification evaluation module is used to identify and evaluate scores according to the daily evaluation results, and the evaluation information storage module is used to store the results of daily evaluations and perform the final evaluation summary. The present invention automatically calculates the evaluation results by manual or AI labeling, which greatly improves the evaluation efficiency. Compared with the traditional manual evaluation calculation method, the AI ​​evaluation system can process more data and information in the same time, thereby meeting the needs of large-scale and efficient evaluation.

[0020] 2. The present invention uses an AI evaluation system through advanced algorithms and models to achieve accurate analysis and evaluation of data. The system can automatically classify, score and evaluate data according to preset rules and standards, thereby reducing interference and errors caused by human factors. This high-precision and high-accuracy evaluation method helps to improve the reliability and fairness of the evaluation results.

[0021] 3. Compared with the traditional manual evaluation method, which requires a large amount of human resources, the AI ​​evaluation system can automatically complete most evaluation tasks. By analyzing historical data and real-time feedback label information, it can automatically calculate and summarize evaluation files and generate evaluation reports at different stages, thereby reducing dependence on manual labor. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a daily evaluation flow chart of the present invention;

[0023] Figure 2 A schematic diagram of the reference standard for daily evaluation scores of the present invention;

[0024] Figure 3 This is a schematic diagram of the daily evaluation flow chart of the present invention and the approval of the leaders;

[0025] Figure 4 A flowchart of the comprehensive evaluation in the daily evaluation record identification evaluation of the present invention;

[0026] Figure 5 This is a flow chart of the monthly assessment summary in the daily assessment record recognition assessment of the present invention;

[0027] Figure 6 The present invention stores daily evaluation records and identifies the evaluation flow chart;

[0028] Figure 7 A schematic diagram of a system interface constructed for an embodiment of the present invention;

[0029] Figure 8 A schematic diagram of the evaluation results of the pending review of the embodiment of the present invention;

[0030] Fig. 9 A schematic diagram showing the review and evaluation results of an individual according to an embodiment of the present invention;

[0031] Fig.10 A schematic diagram of storing daily evaluation records and identifying evaluations in an embodiment of the present invention;

[0032] Fig.11 A schematic diagram of a monthly assessment summary stored in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The present invention is further described in detail below with reference to the accompanying drawings.

[0034] The intelligent automatic evaluation system for smart prisons includes a daily evaluation module, an evaluation record recognition and evaluation module and an evaluation information storage module, wherein the daily evaluation module, the evaluation record recognition and evaluation module and the evaluation information storage module are connected in sequence. The daily evaluation module is used to evaluate the abnormal behaviors of the supervised persons discovered by AI calculations and manual inspections to form the final daily evaluation results; the evaluation record recognition and evaluation module is used to perform recognition and evaluation scoring according to the daily evaluation results; and the evaluation information storage module is used to store the results of daily evaluations and make final evaluation summaries.

[0035] An evaluation method for an intelligent automatic evaluation system for a smart prison comprises the following steps:

[0036] Step 1: The daily evaluation module evaluates the abnormal behaviors of supervised persons discovered by AI calculations and manual inspections to form the final daily evaluation results.

[0037] like Figure 1 As shown in the figure, the daily evaluation module uses AI algorithm (front-end device intelligent algorithm or back-end intelligent behavior analysis algorithm) to analyze the video in the cell, detects abnormal behavior of the supervised persons and pushes the alarm management service, which pushes it to the supervision super brain service after receiving it. The supervision super brain service records it in the daily evaluation of the supervised persons according to the pre-set rules; at the same time, the patrol post manually marks the abnormal behavior of the supervised persons during the online patrol, and the generated abnormal behavior label of the supervised persons will also be recorded in the daily evaluation.

[0038] For example, when the supervised personnel are on duty, if the face scanning on-duty personnel position does not identify the pre-scheduled on-duty personnel or other identified personnel, and if they are not on duty for more than 5 times in a week, they will be automatically identified as on-duty missing tags; the intelligent analysis server can also analyze the single-person on-duty off-duty algorithm, calculate the current on-duty personnel based on the alarm time and the shift plan, and if they are not on duty for more than 5 times in a week, they will also be automatically identified as on-duty missing tags. Specific rules can be configured according to whether the on-site equipment supports face recognition or intelligent analysis.

[0039] like Figure 2 As shown, each label has a corresponding bonus or deduction score. The supervised person has a basic score. The scoring model is "basic score + bonus score - deduction score". The total score per month is the assessment score of the criminal suspect or defendant for that month. The monthly average score is calculated as the total score during the assessment period = the total number of months of full-term assessment.

[0040] Based on the actual situation, the police officers at the disciplinary post evaluate the abnormal behaviors of the supervised persons found by AI calculations and manual inspections, form the final daily evaluation results, and push them to the leaders for subsequent approval;

[0041] like Figure 3 As shown, the leader approves the daily evaluation records pushed by the disciplinary post, and the supervised person confirms the approval on the in-cell screen; if the supervised person does not apply for reconsideration, the daily evaluation information will be recorded in the supervised person's file; if the supervised person applies for reconsideration, the leader needs to conduct reconsideration and approval. After approval, the daily evaluation information will be recorded in the supervised person's file, and daily evaluation related documents will be generated simultaneously.

[0042] Step 2: The evaluation record recognition evaluation module is used to perform recognition evaluation scoring based on daily evaluation results.

[0043] The evaluation record recognition module recognizes and evaluates the daily evaluation records of the supervised persons in the cells under its jurisdiction for a period of time. The evaluation categories include: comprehensive evaluation, monthly assessment summary, and evaluation recognition. After the evaluation is approved, the corresponding electronic document record is generated and recorded in the supervised person's file.

[0044] like Figure 4 As shown, the disciplinary post screens the daily evaluation records and identification evaluation categories of the supervised personnel over a period of time, generates the identification evaluation records, and the leader conducts the final review, adds approval opinions, and generates the identification evaluation of the supervised personnel.

[0045] like Figure 5 As shown, the evaluation record recognition and evaluation module simultaneously generates an electronic evaluation document; at the same time, the evaluation content is formulated, the supervised person checks the formula table on the prison screen, and the correctional police officer queries the supervised person's review record, and the recognition evaluation and evaluation formula are simultaneously generated to generate evaluation-related documents.

[0046] Step 3: The evaluation information storage module stores the daily evaluation results for identification and evaluation.

[0047] like Figure 6 As shown, police officers can choose to create a detention performance file for the supervised person. After the file is created, it will automatically collect the supervised person's basic information, daily evaluation records, identification evaluation records, evaluation formula records, and the corresponding relevant documents in the record.

[0048] Relevant documents include basic information, daily evaluation record form, recognition evaluation record form and evaluation publicity reference form linked to recognition evaluation record form.

[0049] Among them, basic information includes residence permit and arrest warrant.

[0050] The daily evaluation record form includes review documents, evaluation approval and bonus documents, and evaluation deduction and bonus documents.

[0051] The recognition and evaluation record form includes a comprehensive evaluation document, a monthly assessment summary and a performance recognition document.

[0052] The evaluation publicity reference table is associated with the evaluation record table and includes the evaluation publicity documents.

[0053] According to the above-mentioned intelligent automatic evaluation system and method for smart prisons, the construction of a corresponding system has verified the effect of the present invention.

[0054] like Figure 7 As shown, the intelligent automatic evaluation system for smart prisons automatically filters out personnel who need to be evaluated, and can register evaluations for different personnel in the form of tasks. When registering, you can see the label information and label scores under the corresponding label category, and can automatically capture the corresponding evaluation evidence. After registration, it is automatically saved in the evaluation record, and the system automatically calculates according to the evaluation registration information.

[0055] like Figure 8 As shown, after the intelligent automatic evaluation system for smart prisons calculates the evaluation results, the information is automatically transferred to the leader for approval, and a record to be approved is generated. An approval report is automatically generated during the approval process, and the leader can directly perform the approval operation, eliminating the step of manually printing the report.

[0056] like Fig. 9 As shown, after the evaluation information of the intelligent automatic evaluation system for smart prisons is registered and confirmed by multiple police officers, a written report of the corresponding category is automatically generated without the need for manual creation and uploading.

[0057] like Fig.10 and Fig.11As shown, the intelligent automatic evaluation system for smart prisons automatically collects and analyzes relevant valid data after the evaluation, and directly summarizes it into the corresponding detention performance file. The file information of the corresponding personnel can be quickly found and printed.

[0058] According to the above embodiments, the present invention can automatically calculate the evaluation results through manual or AI labeling, which greatly improves the evaluation efficiency. Compared with the traditional manual evaluation calculation method, the AI ​​evaluation system can process more data and information in the same time, thereby meeting the large-scale and efficient evaluation needs.

[0059] The present invention uses an AI evaluation system through advanced algorithms and models to achieve accurate analysis and evaluation of data. The system can automatically classify, score and evaluate data according to preset rules and standards, thereby reducing interference and errors caused by human factors. This high-precision and high-accuracy evaluation method helps to improve the reliability and fairness of the evaluation results.

[0060] Compared with the traditional manual evaluation method, which requires a large amount of human resources, the AI ​​evaluation system can automatically complete most evaluation tasks. By analyzing historical data and real-time feedback label information, it automatically calculates and summarizes evaluation files and generates evaluation reports at different stages, thereby reducing dependence on manual labor.

[0061] It should be emphasized that the embodiments described in the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation manner. Any other implementation manners derived by those skilled in the art based on the technical solution of the present invention also fall within the scope of protection of the present invention.

Claims

1. Intelligent automatic evaluation system for smart prisons, characterized by: It includes a daily evaluation module, an evaluation record identification evaluation module and an evaluation information storage module, wherein the daily evaluation module, the evaluation record identification evaluation module and the evaluation information storage module are connected in sequence. The daily evaluation module is used to evaluate the abnormal behaviors under supervision discovered by AI calculations and manual inspections to form the final daily evaluation results; the evaluation record identification evaluation module is used to identify and evaluate scores based on the daily evaluation results, and the evaluation information storage module is used to store the results of daily evaluations and make final evaluation summaries.

2. An evaluation method for the intelligent automatic evaluation system for smart prisons as claimed in claim 1, characterized in that: The following steps are involved: Step 1: The daily evaluation module evaluates the abnormal behaviors of supervised personnel found by AI calculations and manual inspections to form the final daily evaluation results; Step 2: The evaluation record recognition evaluation module performs recognition evaluation scoring based on the daily evaluation results; Step 3: The evaluation information storage module is used to store the results of daily evaluations and to make final evaluation summaries.

3. The evaluation method for the intelligent automatic evaluation system for smart prisons according to claim 2 is characterized by: The specific implementation method of step 1 is: using AI algorithm to analyze the video in the cell, detecting abnormal behavior of the supervised person and pushing it to the alarm management service, which pushes it to the supervision super brain service after receiving it, and the supervision super brain service records it in the daily evaluation of the supervised person according to the pre-set rules; at the same time, the patrol post manually marks the abnormal behavior of the supervised person during the online patrol, and the generated abnormal behavior label of the supervised person will also be recorded in the daily evaluation. The police at the disciplinary post evaluate the abnormal behavior of the supervised person found by AI calculation and manual patrol according to the actual situation, form the final daily evaluation result, and push it to the leader for subsequent approval; Leaders review and approve daily evaluation records pushed by the disciplinary post, and those approved are confirmed by the supervised persons on the in-cell screen; if the supervised persons do not apply for reconsideration, the daily evaluation information will be recorded in the supervised persons' files; if the supervised persons apply for reconsideration, the leader needs to conduct reconsideration and approval, and after approval, the daily evaluation information will be recorded in the supervised persons' files, and daily evaluation-related documents will be generated simultaneously.

4. The evaluation method for the intelligent automatic evaluation system for smart prisons according to claim 1 is characterized by: The specific implementation method of step 2 is: the disciplinary post screens the daily evaluation records and identification evaluation categories of the supervised persons for a period of time, generates the identification evaluation records, the leader of the station conducts the final review, adds approval opinions, generates the identification evaluation of the supervised persons, and simultaneously generates an electronic evaluation document; at the same time, the evaluation content is formulated, the supervised persons check the formula table on the screen in the prison, the disciplinary police check the supervised persons' reading records, and the identification evaluation and evaluation formula are used to simultaneously generate evaluation-related documents.

5. The evaluation method for the intelligent automatic evaluation system for smart prisons according to claim 1 is characterized by: The specific implementation method of step 3 is: the police can choose to create a detention performance file for the supervised person. After the file is created, it will automatically collect the supervised person's basic information, daily evaluation records, identification evaluation records, evaluation formula records and corresponding relevant documents in the records.

Images