This paper discusses the automatic detection of mura, which has been a long‐standing challenge in the display industries. Using a dataset of 8000 images of OLED (Organic Light Emitting Diode) displays including four different types of mura, we found that a CNN (Convolutional Neural Network) having four or five sets of convolution and max‐pooling layers can detect mura with the accuracy more than 0.8. To improve detection of low contrast mura, we employed a contrast‐enhancement method and a subspace‐method, and the CNN accuracy improved to 0.868, close to the human visible test. Furthermore, the implementation of an automatic in‐line mura‐detection system using the proposed model is also discussed.

This paper discusses the automatic detection of mura, non-uniformity of brightness or color, which has been a long-standing challenge in the display industries. Our purpose is to develop a method using machine learning, which automatically detects and classifies mura in the front-end process. This will enable prompt feedback to the manufacturing process and contribute to improvement of the productivity. We propose "Progressive Hybrid model," which is based on the human visual perception and consists of a multiclass CNN (Convolutional Neural Network), a 2-class residual neural network, and a 2-class CNN. The two 2-class models based on the subspace method to reproduce the boundary-samples in the human visible test are for accurate classification between Normal displays and weak mura. To reproduce the appropriate