Dielectrophoresis-Assisted 3D LC-Oscillator Array in Complementary Metal Oxide Semiconductor Image Senser for Label-Free and Damage Detection of Ancient Building

Abstract

In order to improve the efficiency and accuracy of damage detection of ancient buildings, a dielectrophoresis-assisted 3D LC-oscillator array in CMOS image senser for label-free and damage detection of ancient building is proposed to identify damage areas and achieve pixel-level semantic segmentation. The Grid-Deeplab model is used to model the sub-regions of the damaged image with different importance features. The model has the ability to distinguish the effective area of the image, thereby significantly improve the efficiency and accuracy of the damage detection model. Using the mean intersection over union as the evaluation standard, the proposed Grid-Deeplab model is tested through the data set with the existing U-Net, SegNet, FCN and Deeplab models. The results show that the mean intersection over union of the Grid-Deeplab optimization model on the test set reaches 0.77, and the recognition accuracy and training efficiency of the model are superior to other existing models.