Abstract A key task in computer vision with many applications is object detection. This paper describes an object detection system based on CNN and assesses how well it performs in the particular setting of Iraq. The suggested system makes use of video inputs, breaks them down into individual image frames, and applies pre-processing methods to improve the image quality. To extract significant features, the Sobel operator is used for edge detection and shape identification. A CNN network is created and trained to identify items in the image frames, such as automobiles, people, and buses. Metrics including accuracy, precision, recall, and F1 score are used to assess the system's performance. According to the data, the CNN-based system achieves an F1 score of 91% and scores 92% accurate, 89% precise, 94% recall, and 94% recall. These results demonstrate how well the suggested system performs in the item detection task in the particular research area of Iraq. The CNN-based system outperforms the Random Forest method in terms of accuracy, recall, and F1 score, as shown by a comparison with it. The findings of this study have practical applications in several areas in Iraq, including surveillance, traffic monitoring, and urban planning, and they expand object detection algorithms.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|