scispace - formally typeset
Search or ask a question
Author

Junchao Zhang

Bio: Junchao Zhang is an academic researcher from Central South University. The author has contributed to research in topics: Interpolation & Cardinal point. The author has an hindex of 13, co-authored 44 publications receiving 471 citations. Previous affiliations of Junchao Zhang include University of Arizona & Chinese Academy of Sciences.

Papers
More filters
Journal ArticleDOI
TL;DR: A new interpolation method for DoFP polarimeters is presented by using intensity correlation to detect edges and then implement interpolation along edges, which can achieve better visual effects and a lower RMSE than other methods.
Abstract: Division of focal plane (DoFP) polarimeters operate by integrating micro-polarizer elements with a focal plane. These polarization imaging sensors reduce spatial resolution output and each pixel has a varying instantaneous field of view (IFoV). These drawbacks can be mitigated by applying proper interpolation methods. In this paper, we present a new interpolation method for DoFP polarimeters by using intensity correlation. We employ the correlation of intensity measurements in different orientations to detect edges and then implement interpolation along edges. The performance of the proposed method is compared with several previous methods by using root mean square error (RMSE) comparison and visual comparison. Experimental results showed that our proposed method can achieve better visual effects and a lower RMSE than other methods.

81 citations

Journal ArticleDOI
TL;DR: A new approach is proposed, where the task of phase unwrapping is transferred into a multi-class classification problem and an efficient segmentation network is introduced to identify classes and a noise-to-noise denoised network is integrated to preprocess noisy wrapped phase.
Abstract: The interferometry technique is commonly used to obtain the phase information of an object in optical metrology. The obtained wrapped phase is subject to a 2π ambiguity. To remove the ambiguity and obtain the correct phase, phase unwrapping is essential. Conventional phase unwrapping approaches are time-consuming and noise sensitive. To address those issues, we propose a new approach, where we transfer the task of phase unwrapping into a multi-class classification problem and introduce an efficient segmentation network to identify classes. Moreover, a noise-to-noise denoised network is integrated to preprocess noisy wrapped phase. We have demonstrated the proposed method with simulated data and in a real interferometric system.

78 citations

Journal ArticleDOI
TL;DR: By introducing the variable PRF, the proposed algorithm is equivalent to complete the complex signal processing steps in the radar signal transmission stage, which can greatly improve the efficiency of real-time imaging.
Abstract: Spaceborne synthetic aperture radar (SAR) real-time imaging is especially important for disaster emergencies and real-time monitoring applications with highly desired real-time requirements. Therefore, the continuous improvement of real-time imaging efficiency is an important development trend. At present, traditional real-time imaging algorithms based on constant pulse repetition frequency (PRF) have low accuracy when processing spaceborne SAR data with nonlinear trajectory. For this problem, the existing methods usually introduce some complex signal processing steps, such as scaling or interpolation processing, to improve the accuracy of the real-time imaging, but this will reduce its efficiency. Therefore, this article proposes a new real-time imaging algorithm based on variable PRF for nonlinear trajectory spaceborne SAR. By introducing the variable PRF, the proposed algorithm is equivalent to complete the complex signal processing steps in the radar signal transmission stage, which can greatly improve the efficiency of real-time imaging. Simulation experiments verify the effectiveness of the algorithm.

64 citations

Journal ArticleDOI
TL;DR: The results validate that the proposed algorithm using residual interpolation can give state-of-the-art performance over several previously published interpolation methods, namely bilinear, bicubic, spline and gradient-based interpolation.
Abstract: Division of focal plane (DoFP) polarization image sensors capture polarization properties of light at every imaging frame. However, these imaging sensors capture only partial polarization information, resulting in reduced spatial resolution output and a varying instantaneous field of overview (IFoV). Interpolation methods are used to reduce the drawbacks and recover the missing polarization information. In this paper, we propose residual interpolation as an alternative to normal interpolation for division of focal plane polarization image sensors, where the residual is the difference between an observed and a tentatively estimated pixel value. Our results validate that our proposed algorithm using residual interpolation can give state-of-the-art performance over several previously published interpolation methods, namely bilinear, bicubic, spline and gradient-based interpolation. Visual image evaluation as well as mean square error analysis is applied to test images. For an outdoor polarized image of a car, residual interpolation has less mean square error and better visual evaluation results.

57 citations

Journal ArticleDOI
TL;DR: Experimental results show that the proposed convolutional neural network to address the image demosaicing issue outperforms other state-of-the-art methods by a large margin in terms of quantitative measures and visual quality.
Abstract: We propose a polarization demosaicing convolutional neural network to address the image demosaicing issue, the last unsolved issue in microgrid polarimeters. This network learns an end-to-end mapping between the mosaic images and full-resolution ones. Skip connections and customized loss function are used to boost the performance. Experimental results show that our proposed network outperforms other state-of-the-art methods by a large margin in terms of quantitative measures and visual quality.

48 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: In this article, the electron beam assisted evaporation technique is analyzed along with other methods operating at oblique angles, including, among others, magnetron sputtering and pulsed laser or ion beam-assisted deposition techniques.

537 citations

Journal ArticleDOI
TL;DR: Deep learning-enabled optical metrology is a kind of data-driven approach, which has already provided numerous alternative solutions to many challenging problems in this field with better performances as discussed by the authors .
Abstract: Abstract With the advances in scientific foundations and technological implementations, optical metrology has become versatile problem-solving backbones in manufacturing, fundamental research, and engineering applications, such as quality control, nondestructive testing, experimental mechanics, and biomedicine. In recent years, deep learning, a subfield of machine learning, is emerging as a powerful tool to address problems by learning from data, largely driven by the availability of massive datasets, enhanced computational power, fast data storage, and novel training algorithms for the deep neural network. It is currently promoting increased interests and gaining extensive attention for its utilization in the field of optical metrology. Unlike the traditional “physics-based” approach, deep-learning-enabled optical metrology is a kind of “data-driven” approach, which has already provided numerous alternative solutions to many challenging problems in this field with better performances. In this review, we present an overview of the current status and the latest progress of deep-learning technologies in the field of optical metrology. We first briefly introduce both traditional image-processing algorithms in optical metrology and the basic concepts of deep learning, followed by a comprehensive review of its applications in various optical metrology tasks, such as fringe denoising, phase retrieval, phase unwrapping, subset correlation, and error compensation. The open challenges faced by the current deep-learning approach in optical metrology are then discussed. Finally, the directions for future research are outlined.

165 citations

Journal ArticleDOI
TL;DR: This paper compares the performance of the standard state-of-the-art object detectors that were retrained on a dataset of thermal images extracted from videos that simulate illegal movements around the border and in protected areas and presents the results of the recognition of humans and animals in thermal images.
Abstract: Global terrorist threats and illegal migration have intensified concerns for the security of citizens, and every effort is made to exploit all available technological advances to prevent adverse events and protect people and their property. Due to the ability to use at night and in weather conditions where RGB cameras do not perform well, thermal cameras have become an important component of sophisticated video surveillance systems. In this paper, we investigate the task of automatic person detection in thermal images using convolutional neural network models originally intended for detection in RGB images. We compare the performance of the standard state-of-the-art object detectors such as Faster R-CNN, SSD, Cascade R-CNN, and YOLOv3, that were retrained on a dataset of thermal images extracted from videos that simulate illegal movements around the border and in protected areas. Videos are recorded at night in clear weather, rain, and in the fog, at different ranges, and with different movement types. YOLOv3 was significantly faster than other detectors while achieving performance comparable with the best, so it was used in further experiments. We experimented with different training dataset settings in order to determine the minimum number of images needed to achieve good detection results on test datasets. We achieved excellent detection results with respect to average accuracy for all test scenarios although a modest set of thermal images was used for training. We test our trained model on different well known and widely used thermal imaging datasets as well. In addition, we present the results of the recognition of humans and animals in thermal images, which is particularly important in the case of sneaking around objects and illegal border crossings. Also, we present our original thermal dataset used for experimentation that contains surveillance videos recorded at different weather and shooting conditions.

130 citations

Journal ArticleDOI
Akira Saito1
TL;DR: The basic principles of natural photonic materials, the ideas developed from these principles, the directions of applications and practical industrial realizations are presented by summarizing the recent research results.

95 citations

Journal ArticleDOI
TL;DR: Deep learning-enabled optical metrology is a kind of data-driven approach, which has already provided numerous alternative solutions to many challenging problems in this field with better performances as discussed by the authors .
Abstract: Abstract With the advances in scientific foundations and technological implementations, optical metrology has become versatile problem-solving backbones in manufacturing, fundamental research, and engineering applications, such as quality control, nondestructive testing, experimental mechanics, and biomedicine. In recent years, deep learning, a subfield of machine learning, is emerging as a powerful tool to address problems by learning from data, largely driven by the availability of massive datasets, enhanced computational power, fast data storage, and novel training algorithms for the deep neural network. It is currently promoting increased interests and gaining extensive attention for its utilization in the field of optical metrology. Unlike the traditional “physics-based” approach, deep-learning-enabled optical metrology is a kind of “data-driven” approach, which has already provided numerous alternative solutions to many challenging problems in this field with better performances. In this review, we present an overview of the current status and the latest progress of deep-learning technologies in the field of optical metrology. We first briefly introduce both traditional image-processing algorithms in optical metrology and the basic concepts of deep learning, followed by a comprehensive review of its applications in various optical metrology tasks, such as fringe denoising, phase retrieval, phase unwrapping, subset correlation, and error compensation. The open challenges faced by the current deep-learning approach in optical metrology are then discussed. Finally, the directions for future research are outlined.

95 citations