Advance Search
中文

2021 Vol. 43, No. 4

column
  Previous Issue | Next Issue 
Display Method:
Materials & Devices
Analysis of Interface Control Methods for InAs/GaSb Type-Ⅱ Superlattice Materials Grown by MBE
REN Yang, LI Junbin, QIN Gang, YANG Jin, LI Yanhui, ZHOU Xuchang, YANG Chunzhang, CHANG Chao, KONG Jincheng, LI Dongsheng
2021, 43(4): 301-311.
Abstract HTML (231) PDF(138)
Abstract:
This article systematically introduces interface control methods for the MBE growth of InAs/GaSb type-Ⅱ superlattice materials, including the interrupted growth epitaxy method, migration-enhanced epitaxy, V group element soak method, and bulk material growth method. The short-wavelength (mid-wavelength) InAs/GaSb superlattice material interface adopts a mixed-like interface, and the control method is mainly the interrupted growth epitaxy method, the long-wavelength (very long-wavelength) superlattice material interface adopts the InSb-like interface, and the control method adopts the migration-enhanced epitaxy (MEE) or Sb soak method combined with bulk material growth. The basis for selecting the interface type of InAs/GaSb superlattice material is discussed and analyzed, and the specific implementation theory of interface control is briefly described, along with the selection of interface types and control methods of superlattice materials in different infrared detection wavelength bands by related research institutions. To effectively improve the stress compensation effect of the interface layer, the interface structure epitaxial growth process design, that is, the experimental design of different shutter sequences based on the interface control method, was used. This is of great significance for the optimization of the crystal quality and device performance of long-wave, very long-wave, and two-color (even multi-color) superlattice materials.
Research Development of Infrared Stealth Materials
SHEN Yulian, LI Chunhai, GUO Shaoyun, CHEN Rong
2021, 43(4): 312-323.
Abstract HTML (646) PDF(427)
Abstract:
With the rapid development of infrared detection technology, the improvement of the infrared stealth capability of military targets has become an urgent problem to be solved, so it is of great significance to study infrared stealth materials. This paper briefly analyzes the stealth mechanism of infrared stealth materials, summarizes the research status of four types of infrared stealth materials in recent years, such as low infrared emissivity materials, temperature control materials, photonic crystals, and intelligent infrared stealth materials, and forecasts the future development trend of infrared stealth materials.
Non-uniformity Correction for Large Format Array Infrared Detectors Based on Regional Correction
ZHANG Mingjie, LI Yan, MA Wenpo, LIU Ziying
2021, 43(4): 324-333.
Abstract HTML (181) PDF(113)
Abstract:
By analyzing the response characteristics of large-format infrared detectors, we found that the response nonlinearity in different areas is caused by the characteristics of the camera itself. After the traditional two-point correction method or nonlinear curve fitting method corrects the large-format array detector, the correction residuals and visual effects are relatively poor. In this study, on the basis of eliminating blind elements, according to the nonlinear response characteristics of the detector, the entire array is divided into eight regions for nonlinear fitting correction, the bias coefficient of each region was corrected, and the non-uniformity correction algorithm of the BP neural network was used to deal with the non-uniformity problem caused by region division. After correction, the residual non-uniformity of each black body temperature point image was on the order of one-thousandth, the spatial noise was already very close to or smaller than the temporal noise, and the local residual non-uniformity reached below 0.002, which was significantly smaller than the temporal noise.
Image Processing & Simulation
A Method of Image Classification for Objects with Camouflaged Color Features
LIU Feng, LI Jiajun, LI Yuhai, GAO Peipei
2021, 43(4): 334-341.
Abstract HTML (144) PDF(34)
Abstract:
To solve the classification difficulty caused by different objects with similar color features in an image, this paper proposes an image classification method for objects with camouflaged color features. To alleviate the difficulty of distinguishing color-camouflaged objects in RGB images, this method not only combines the spatial domain features of the neighborhood pixels and the spectral domain features of the hyperspectral data, which realizes the spatial-spectrum joint feature construction, but also uses the improved LLE(local linear embedding) algorithm to accomplish spectral dimensional reduction. The proposed method uses an active learning capsule network to train a hyperspectral data classifier and classifies objects in the scene. Active learning can label more representative samples through the improved active learning function and realized capsule network training based on a minor sample dataset, which reduces the cost of sample labeling and model training significantly, thereby improving the classification performance of the model. Experiments show that the algorithm proposed in this paper can effectively classify camouflaged targets and other natural targets based on our self-made hyperspectral dataset. The average accuracy of camouflaged targets was 91%, and the average accuracy of all target types was 89.9%.
Infrared Vehicle Target Detection Based on Convolutional Neural Network without Pre-training
CHEN Gao, WANG Weihua, LIN Dandan
2021, 43(4): 342-348.
Abstract HTML (156) PDF(61)
Abstract:
To tackle the over-dependence of convolutional neural network-based target detection algorithms on pre-training weights, especially for target detection of infrared scenarios under data-sparse conditions, the incorporation of attention modules is proposed to alleviate the degradation of detection performance owing to the absence of pre-training. This paper is based on the YOLO v3 algorithm, which incorporates SE and CBAM modules in a network that mimics human attentional mechanisms to recalibrate the extracted features at the channel and spatial levels. Different weights are adaptively assigned to the features according to their importance, which ultimately improves the detection accuracy. On the constructed infrared vehicle target dataset, the attention module significantly improved the detection accuracy of the non-pre-trained convolutional neural network. Furthermore, the detection accuracy of the network incorporating the CBAM module was 86.3 mAP, demonstrating that the attention module can improve the feature extraction ability of the network and free the network from over-reliance on the pretrained weights.
Small Infrared Target Detection Based on Fully Convolutional Network
YANG Qili, ZHOU Binghong, ZHENG Wei, LI Mingtao
2021, 43(4): 349-356.
Abstract HTML (86) PDF(107)
Abstract:
In the field of aerospace research, such as in small celestial body detection, missile guidance, and battlefield reconnaissance, because the target signal is weak, the number of pixels occupied is small, and the target lacks shape structure and texture information, traditional algorithms with manual feature extraction are prone to false alarms, whereas deep learning methods with powerful feature extraction capabilities cannot train tiny targets that lack contour information. In this context, a sliding window sampling training method is adopted, which originates from the idea of nested structures in traditional algorithms based on human visual characteristics. A fully convolutional network using recursive convolutional layers is designed to extend the depth of the network without increasing the training parameters. The multi-branch structure of the network's parallel convolution structure simulates the multi-scale operation of the traditional algorithm, which can enhance the contrast between the target and the background. Additionally, various loss functions are designed to combat the serious imbalance between positive and negative samples. The results show that the algorithm achieves a better detection performance than the traditional algorithms.
Infrared Small Dim Target Detection Based on Local Contrast Mechanism
HAN Jinhui, DONG Xinghao, JIANG Yawei, LI Zhizheng, LIANG Kun, ZHANG Lihong
2021, 43(4): 357-366.
Abstract HTML (132) PDF(105)
Abstract:
A method for infrared (IR) small dim target detection based on a local contrast mechanism is proposed to solve the problem of IR small dim target detection under a complex background and low signal-to-clutter ratio (SCR). A three-layer window consisting of an inner layer, a middle layer, and an outer layer is proposed, so that targets of different scales can be detected using only single-scale calculations. First, the matched filter is applied to the inner layer to enhance the true target purposefully, and the max-close principle is proposed to estimate the background of the outer layer, so that detection becomes easier when the target is near the background edge. Then, the ratio-difference joint local contrast measure is calculated between the enhanced target and the estimated background to enhance the true target and suppress the complex background simultaneously. Finally, an adaptive threshold operation is used to extract the true target. Experimental results show that compared to some existing algorithms, the proposed algorithm can enhance the true target and suppress complex background better, and its principle is simple yet suitable for implementation and can effectively reduce the amount of calculation.
Image Segmentation of Inductors Laser Thermal Imaging Based on Watershed Algorithm
ZHANG Yishu, WANG Xiaona, HOU Dexin, YE Shuliang
2021, 43(4): 367-371.
Abstract HTML (183) PDF(29)
Abstract:
Laser thermography is a new method to detect micro cracks in inductors, but the proximity of inductors in batch automatic detection made it easy to cause positioning errors and false detection results. According to the analysis of image features in the process of laser scanning and thermal imaging detection, a new method was proposed to obtain the temperature gradient map and eliminate its excitation nonuniformity by using the sub maximum filter, and to disconnect the adhesion of samples and remove the interference noise points by using morphological operation. The inductance images were segmented combining distance transformation with watershed algorithm. The experimental results show that the method realizes the automatic separation and extraction of ferrite inductor from image, which lays a good foundation for the separation of ferrite inductors.
An Infrared Image Segmentation Method Based on Improved Lazy Snapping Algorithm
ZHANG Lian, LI Mengtian, YU Songlin, GONG Yu, YANG Hongjie
2021, 43(4): 372-377.
Abstract HTML (115) PDF(38)
Abstract:
Considering that infrared images contain a considerable amount of noise and are of low contrast, an improved lazy snapping(LS) segmentation method combined with fast fuzzy C-means clustering is proposed. Infrared images are pre-segmented using a fast fuzzy C-means clustering algorithm, and the target and background seed points are marked in the image by the morphological skeleton extraction method. The LS algorithm is converted from global segmentation to cluster region segmentation, and an energy function is constructed. The minimum value of the energy function is solved by the minimum cut algorithm, and the segmentation efficiency is improved. The phenomenon of over-segmentation in the image is reduced, the LS algorithm is changed from an interactive algorithm to a non-interactive algorithm. Thus, the automatic segmentation of infrared images is realized, improving the real-time nature of the LS algorithm. By performing segmentation experiments on various infrared images and then comparing the proposed method's performance with that of other segmentation methods, the results show that the improved algorithm has a good segmentation effect and strong robustness.
Fast Restoration Algorithm for Space-variant Defocus Blurred Infrared Images
WANG Chenyue, LEI Xufeng, LI Zemin, YANG Shaoming, HE Yan
2021, 43(4): 378-384.
Abstract HTML (84) PDF(42)
Abstract:
A fast restoration algorithm based on image quality assessment is proposed to improve the quality of space-variant defocus blurred infrared images. First, the defocus image is restored by the truncated constrained least-squares algorithm with different point spread functions to obtain and perform deringing on multiple restored images. Then, the area centered on each pixel in the restored image is evaluated through an image quality assessment, and the images restored with different parameters are combined according to the image quality assessment to obtain the final restored image. Because there is no need to estimate the point spread function of the blurred image, the truncated constrained least-squares algorithm of spatial calculation is used for restoration. The experimental results show that the algorithm proposed in this paper can quickly restore the space-variant defocused blurred infrared image and is much faster than the method based on point spread function estimation.
Systems & Designs
Software Design for Turntable Control System of Small Portable Step Gaze Infrared Search and Track System
QU Zuxin, TIAN Guiping, CHEN Jie
2021, 43(4): 385-390.
Abstract HTML (162) PDF(42)
Abstract:
A turntable motor control system was constructed using domestic MM32F031 as the control chip. The software design for controlling movement was mainly carried out. To realize the different motion modes of the turntable, the motion control function was analyzed. Combined with the software design mode and data communication mode, the software was designed, and the supporting upper computer software was designed. After testing, the software met the system requirements.
Terahertz
Terahertz Image Enhancement Based on Generative Adversarial Network
ZHANG Pengcheng, HE Mingxia, CHEN Shuo, ZHANG Hongzhen, ZHANG Xinxin
2021, 43(4): 391-396.
Abstract HTML (232) PDF(54)
Abstract:
In terahertz scanning imaging, the image contrast is low due to laser power fluctuation and instrument vibration, and the imaging quality needs to be improved. At present, the processing of terahertz image is still in the traditional algorithm stage. In this paper, an image enhancement method based on Generative Adversarial Network is proposed, which includes the idea of deep learning. By introducing blur and noise into the training set image, the mapping relationship between low-quality images and high-quality images is learned and applied to real terahertz images. The experimental results show that, compared with traditional algorithms such as bilateral filtering and non-local mean filtering, this method can significantly improve the image contrast on the basis of improving image details, and has a good visual sense, which provides a new idea for terahertz image enhancement.
IR Applications
Infrared Thermography Low-zero Insulator Identification Based on GWO-SVM
ZHANG Meijin, QU Qiubo
2021, 43(4): 397-402.
Abstract HTML (153) PDF(29)
Abstract:
The accuracy of the diagnosis of degraded insulators is improved to accurately identify low-zero-value insulators in the power grid. A pair of insulator infrared images and a gray wolf optimizer (GWO) optimized binary support vector machine (SVM) classifier is proposed. Low-zero insulators are detected automatically. First, the infrared image of the insulator is enhanced; then, the infrared image is segmented using the Ostu algorithm; and the obtained binary image is subjected to tilt angle correction and cutting to extract the effective region of the insulator string. Finally, the image features are applied to the classification and recognition of vector machines. The experimental results show that the GWO-SVM can identify the low-zero insulator more accurately and effectively than the commonly used grid search (GS) and particle swarm optimization (PSO). Its rate is higher.
Refrigeration
Fractal Characterization of Regenerator of Micro Stirling Coolers
LI Renzhi, CHEN Xiaoping, SUN Hao, LI Haolan
2021, 43(4): 403-408.
Abstract HTML (90) PDF(47)
Abstract:
To investigate the microstructural characteristics of the filling structure in a regenerator, based on the fractal theory of porous media, the mercury intrusion method was used to study the pore distribution and fractal dimension of the regenerator. The regenerator is a key component of a miniature Stirling cooler. The regenerator prepared by filling stainless steel mesh or stainless steel felt is a typical porous medium. The microstructure of the regenerator was tested using the mercury intrusion method, and the pore distribution range inside the regenerator was obtained. Combined with the basic theory of fractal analysis of porous media, the fractal dimension of the regenerator is calculated, which shows that the regenerator has fractal characteristics, and the fractal dimension interval of the regenerator can be obtained.

Copyright © 《Infrared Technology》Editorial Office京ICP备000000000号-1

Supported by: Beijing Renhe Information Technology Co. Ltd