Authors:
Sergey Turko
1
;
Liudmila Burmak
1
;
Ilya Malyshev
1
;
Stanislav Shtykov
1
;
Mikhail Popov
1
;
Pavel Filimonov
2
;
Alexandr Aspidov
2
;
1
and
Andrei Shcherbinin
1
Affiliations:
1
Samsung R&D Institute Russia, 12 Dvintsev Str., Moscow, Russia
;
2
Bauman Moscow State University, 5 2nd Baumanskaya Str., Moscow, Russia
Keyword(s):
Glass Inspection, Optical Inspection, Dark-field Imaging, U-Net, Imbalanced Data, Nested Weights, Semantic Segmentation, Image Processing.
Abstract:
In this paper we address the problem of detection and discrimination of defects on smartphone cover glass. Specifically, scratches and scratch-like defects. An automatic detection system which allows to detect scratches on the whole surface of a smartphone’s cover glass without human participation is developed. The glass sample is illuminated sequentially from several directions using a special ring illumination system and a camera takes a dark-field image at each illumination state. The captured images show a variation of the defect image intensity depending on the illumination direction. We present a pipeline of detecting scratches on images obtained by our system using convolutional neural networks (CNN) and particularly U-net-like architecture. We considered the scratch detection problem as a semantic segmentation task. The novel loss technique for solving the problem of imbalance, sparsity and low representability of data is presented. The proposed technique solves two tasks sim
ultaneously: segmentation and reconstruction of the provided image. Also, we suggest a nested convolution kernels to overcome the problem of overfitting and to extend the receptive field of the CNN without increasing trainable weights.
(More)