Deep Learning for Quality Assurance in Casting Production: ResNet152V2 Approach

Quality control of the final product is crucial in modern production. The feasibility and effectiveness of automating quality inspection operations through image analysis have increased with the development of computer vision and deep learning techniques. This article presents a novel convolutional neural network (CNN) architecture called ResNet152V2 that we believe can be used to examine casting products for quality. Using a dataset of casting product photos, we refine the already-trained ResNet152V2 model to make use of transfer learning. The model can learn discriminative features indicative of product quality because the dataset contains photos of both faulty and non-defective products. Through the use of transfer learning, we leverage the learned representations from a large-scale dataset, reducing the amount of training data and computer power required. We demeanor wide experiments to show the efficacy of the suggested method, assessing the model's performance in terms of F1 score, accuracy, precision, and recall. The experimental results reveal how fine the ResNet152V2 prototypical can detect faulty casting products, which can help manufacturing environments run more effective quality control procedures. With possible advantages like higher production efficiency, lower inspection costs, and better product quality, this research advances automated quality inspection systems.