According to a classification of central nervous system tumors by the World Health Organization, diffuse gliomas are classified into grade 2, 3, and 4 gliomas in accordance with their aggressiveness. To quantitatively evaluate a tumor’s malignancy from brain magnetic resonance imaging, this study proposed a computer-aided diagnosis (CAD) system based on a deep convolutional neural network (DCNN). Gliomas from a multi-center database (The Cancer Imaging Archive) composed of a total of 30 grade 2, 43 grade 3, and 57 grade 4 gliomas were used for the training and evaluation of the proposed CAD. Using transfer learning to fine-tune AlexNet, a DCNN, its internal layers, and parameters trained from a million images were transferred to learn how to differentiate the acquired gliomas. Data augmentation was also implemented to increase possible spatial and geometric variations for a better training model. The transferred DCNN achieved an accuracy of 97.9% with a standard deviation of ±1% and an area under the receiver operation characteristics curve (Az) of 0.9991 ± 0, which were superior to handcrafted image features, the DCNN without pretrained features, which only achieved a mean accuracy of 61.42% with a standard deviation of ±7% and a mean Az of 0.8222 ± 0.07, and the DCNN without data augmentation, which was the worst with a mean accuracy of 59.85% with a standard deviation ±16% and a mean Az of 0.7896 ± 0.18. The DCNN with pretrained features and data augmentation can accurately and efficiently classify grade 2, 3, and 4 gliomas. The high accuracy is promising in providing diagnostic suggestions to radiologists in the clinic.

Based on a line scan imager, we propose anew defect inspection system to precisely identify defects with size of greater than criteria from chip backside. Only a corresponding raw image is acquiredfor each chip,and using anopto-mechanical device with very low geometry distortion to make it feasible to neglectprocesses of image calibration and mosaic. Defect inspection is based on a binary chip edge image and a novel methodof using edge pixels statistic. Thus defect size of crack, chipping and glue can be inspected quantitatively and efficiently.From test results by integrating inspection system with experimental and operational chip sorters, our system has capabilities of maximal chip moving speed of0.7 m/sec (5μm resolution) andinspecting chip defects with size accuracyof less than 1.0 pixel. Proposed system has features of automatically acquiring chip images during packaging process, improving performance of image processing and defects inspection. Moreover, manpower, time and cost for chip defect inspection manually can be saved. Chip quality control can be improved as well comparing with the one by manual way.Developed system has been integrated with operational chip sorters successfully for sale now.

This paper presents a machine vision inspection method for winding high frequency inductors, which affects the reliability and quality of the electronic products. This paper proposes how to quickly and correctly improve the quality of component detection, an important issue for surface mounted device (SMD) inductors manufacturers. SMD components easily damage the phenomenon of the electrode, and the brightness of the brightness of the damaged area of the electrode close to normal, not easy to be precise defect area separated from the electrode area.