Prisoner prison break intention evaluation method based on body language

Abstract

The invention discloses a prisoner prison break intention evaluation method based on a body language. The method comprises the following steps: extracting skeleton information from a personal image ofa prisoner in each frame of image of a monitoring video to obtain a skeleton information sequence; constructing and training an RNN and fuzzy reasoning fused network; and inputting the skeleton information sequence into the constructed RNN and fuzzy reasoning fused network to output the prison break intention of the prisoner. The method has the advantages of being high in accuracy, high in credibility level and good in real-time performance. Limb language features of prisoners are collected through a camera, the requirement for transformation of prison equipment is low, and the situation thatface images are possibly blurred is fully considered. The network combines RNN with fuzzy reasoning, so that temporal information related to the body language can be extracted, and the problems of fuzziness and noise are solved. And a non-contact collection and evaluation method avoids psychological discomfort of the prisoners.

Description

technical field [0001] The invention belongs to the technical field of judicial supervision, and in particular relates to a body language-based evaluation method for prison escape intentions of prisoners. Background technique [0002] The escape behavior of inmates is unpredictable, which is a major problem in risk management and control in prisons. Immediately and accurately grasping the psychological fluctuations of prisoners is the fundamental method to predict and control their behavioral trends. Grasping the thoughts and actions of prisoners is an important means to stick to the bottom line of safety, improve the safety governance system, and create a safe prison. How to use the prison's existing video equipment and transmission system to use AI technology to obtain and evaluate the true psychological status and trends of prisoners is a new demand for risk management and control in prisons. [0003] Generally speaking, risk management and control in prisons in China i...

AI Technical Summary

Problems solved by technology

However, most of these methods rely on subjective judgments, with poor accuracy, high labor costs, and a certain lag in processing time.

A few places have begun to use mobile-based scale assessment tools. In terms of assessment content, there are few studies focusing on dynamic factors such as occupation, personality, psychological status, and mental illness. The collection of dynamic psychological states based on daily life has not yet been seen. to report

Images