A Machine Vision and Integrated Sensors-Based Portable Forensic Field Inspection Device
摘要
This paper introduces a portable forensic field inspection device that leverages machine vision and integrated sensors for real-time data collection and analysis. The device incorporates temperature, humidity, gas sensors, and a high-resolution imaging sensor to capture comprehensive on-site data. A machine learning model, enhanced with the Convolutional Block Attention Module (CBAM), is employed to analyze the collected images, improving the accuracy and efficiency of determining the cause of death. The study details the design of the device and the data processing methods, including chrominance enhancement, data annotation, and augmentation techniques. Experimental results validate the high precision of the sensors and the superior performance of the machine learning model, demonstrating significant improvements in forensic fieldwork.