SURFACE DEFECTS DETECTION BASED ON ADAPTIVE MULTISCALE IMAGE COLLECTION AND CONVOLUTIONAL NEURAL NETWORKS

Abstract

Surface flaw inspection is of great importance for quality control in the field of manufacture. In this paper, a novel surface flaw inspection algorithm is proposed based on adaptive multiscale image collection (AMIC) using convolutional neural networks. First, the inspection networks are pretrained with ImageNet data set. Second, the AMIC is established, which consists of adaptive multiscale image extraction and with-contour local extraction from training images. Through the AMIC, the training data set is greatly augmented, and labels of images can be accomplished automatically without artificial consumption. Then, transfer learning is performed with the AMIC established from training data set. Finally, an automatic surface flaw inspection instrument for large-volume metal components embedded with the proposed inspection algorithm is designed. Experiments with small metal components are performed to analyze the influence of parameters, and comparative experiments are carried out. The inspecting precisions for indentation, scratch, and pitted surface of the proposed method are 97.3%, 99.5%, and 100%, respectively. The experimental results demonstrate the effectiveness of the proposed method in the detection of various surface flaws.

Existing System

Support vector machine (SVM), K nearest neighbor

Proposed System

In this proposed system, a surface flaw inspection and classification framework for small metal components is presented. The whole inspection algorithm consists of two main processes: training process and detection process. CNNs and transfer learning are used in training processing for feature extraction. Meanwhile, AMIC is proposed as an augmentation method for training images. In detection processing, images acquired in real time are preprocessed and cropped; then, these image blocks are inspected and classified by the trained CNNs.

Architecture

BLOCK DIAGRAM