An Automatic Forensic Injury Annotation Method and System Based on Image Segmentation Technology
By using an automatic injury labeling method based on image segmentation technology, the automatic labeling device captures video images of wounds and predicts the best shooting angle, solving the problem of low efficiency in manually filling in injury results by forensic doctors, and realizing rapid and accurate injury level assessment and labeling.
Patent Information
- Application Number
- CN202411127919.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2025-11-14
- Estimated Expiration
- 2044-08-16
AI Technical Summary
The current method of injury investigation requires forensic doctors to manually fill in the results, which is inefficient and cannot meet the actual needs.
An automatic injury labeling method for forensic injury identification based on image segmentation technology is adopted. Video images of the wound area are captured by an automatic labeling device, the external shape data of the wound is extracted using image segmentation technology, and the target display posture is predicted by combining the posture prediction model to determine the best shooting angle. The injury level is then automatically labeled in the high-definition image.
It enables rapid and automatic assessment and labeling of injury severity, greatly reducing the workload of forensic doctors, improving the efficiency of injury identification, and ensuring the accuracy of injury severity rating.