A Systematic Review: Development of AI Based Computer Vision Scrap Sorting System for Metal Scrap

The metals recycling industry has undergone a revolution thanks to the quick development of computer vision and artificial intelligence (AI), which have made it possible to sort metal trash using efficient and automated techniques. The existence of pollutants, the requirement for high-speed processing without sacrificing precision, and variations in scrap form, size, and texture are some of the ongoing difficulties. Though they provide strong feature extraction capabilities, current methods like Convolutional Neural Networks (CNNs), Scale-Invariant Feature Transform (SIFT), and Histogram of Oriented Gradients (HOG) frequently necessitate substantial preprocessing and sizable, precisely labeled datasets in order to integrate them with classification models like Support Vector Machines (SVM) and Artificial Neural Networks (ANN). Furthermore, scalability and resilience in real-world scenarios continue to be major challenges. An AI-driven computer vision system is proposed in this work for identifying and sorting different kinds of metal trash, such as alloys made of copper, iron, and aluminium. Utilizing sophisticated feature extraction techniques and combining SVM and ANN models, the system seeks to maximize classification performance while tackling issues with operational effectiveness and dataset variability. YOLO LabelImg makes annotations easier by producing accurate bounding boxes, which are essential for tasks involving object identification and categorization. The study emphasizes the value of integrating models to increase accuracy and investigates how AI-powered solutions might raise the efficiency, dependability, and affordability of applications for sorting metal waste.