研究目的
To detect suspicious devices in cars using X-ray image processing technique to prevent terrorism and enhance security by reducing detection time and improving accuracy.
研究成果
The prototype system achieved 80% accuracy in detecting weapons using the Canny Edge Detector and template matching algorithms. It demonstrates potential for reducing security risks by automating detection, but improvements are needed for handling unclear images and reducing false positives.
研究不足
The system may fail to detect weapons if images are unclear or if there are objects that resemble weapons, leading to errors. The use of internet-sourced images instead of real X-ray images may limit realism and applicability.
1：Experimental Design and Method Selection:
The study uses an image processing system based on human vision principles, employing the Canny Edge Detector algorithm for edge detection, which includes smoothing with a Gaussian filter, gradient calculation, non-maxima suppression, and double thresholding. Template matching is used for weapon recognition.
2：Sample Selection and Data Sources:
Images were obtained from the internet due to the lack of an X-ray camera. A database of weapon images is used for training and comparison.
3：List of Experimental Equipment and Materials:
No specific equipment or materials are listed in the paper; images are sourced from the internet.
4：Experimental Procedures and Operational Workflow:
The process involves pre-processing images, converting to grayscale and binary, edge detection using Canny algorithm, weapon recognition via template matching, and sending notifications if suspicious objects are detected.
5：Data Analysis Methods:
Accuracy is measured by comparing detected images with weapon images in the database, with results showing 80% accuracy.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
