Transparent Object Reconstruction Based on Compressive Sensing and Super-Resolution Convolutional Neural Network

作     者：Anumol MATHAI Li MENGDI Stephen LAU Ningqun GUO Xin WANG Anumol MATHAI;Li MENGDI;Stephen LAU;Ningqun GUO;Xin WANG

作者机构：School of EngineeringMonash University MalaysiaSelangor 47500Malaysia College of Optoelectronic EngineeringChangchun University of Science and TechnologyJilin 130022China 

出 版 物：《Photonic Sensors》 (光子传感器（英文版）)

年 卷 期：2022年第12卷第4期

页      面：24-35页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 080202[工学-机械电子工程] 08[工学] 081104[工学-模式识别与智能系统] 0804[工学-仪器科学与技术] 0802[工学-机械工程] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0702[理学-物理学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This research was funded by the Ministry of Higher Education Malaysia(Grant No.Grant FRGS/1/2020/ICT02/MUSM/02/1) 

主　　题：Transparent object imaging single-pixel imaging compressive sensing total-variation minimization SRCNN algorithm 

摘      要：The detection and reconstruction of transparent objects have remained challenging due to the absence of their features and variations in the local features with variations in *** this paper,both compressive sensing(CS)and super-resolution convolutional neural network(SRCNN)techniques are combined to capture transparent *** the proposed method,the transparent object’s details are extracted accurately using a single pixel detector during the surface *** resultant images obtained from the experimental setup are low in quality due to speckles and deformations on the ***,the implemented SRCNN algorithm has obviated the mentioned drawbacks and reconstructed images visually *** developed algorithm locates the deformities in the resultant images and improves the image ***,the inclusion of compressive sensing minimizes the measurements required for reconstruction,thereby reducing image post-processing and hardware requirements during network *** result obtained indicates that the visual quality of the reconstructed images has increased from a structural similarity index(SSIM)value of 0.2 to *** this work,we demonstrate the efficiency of the proposed method in imaging and reconstructing transparent objects with the application of a compressive single pixel imaging technique and improving the image quality to a satisfactory level using the SRCNN algorithm.