This paper compares four different fabrication methods for mercury cadmium telluride (HgCdTe) p-on-n devices. Among these methods, vertical liquid-phase epitaxy (VLPE) stands out because of its unique advantages, particularly the high activation rate of in situ arsenic (As) dopants. VLPE is an essential approach for producing high-performance p-on-n double heterojunction devices. This paper reviews the research progress, both domestically and internationally, covering material growth, device processes, and performance. The discrepancies between domestic and foreign research are discussed, and the key challenges and technical bottlenecks hindering VLPE technology development are identified. Several solutions have been proposed to solve this problem. This study provides insights into the future trends of VLPE technology for p-on-n heterojunction devices, which hold significant promise in semiconductor devices.
Shortwave infrared imaging is a popular topic worldwide. By receiving shortwave infrared radiation for detection and imaging, more information about the target objects can be obtained, compensating for the shortage of visible light imaging to achieve full-band imaging. Based on the optical characteristics, imaging principle, and optical system structure design of shortwave infrared optical imaging, this study compares the advantages and disadvantages of shortwave infrared imaging with visible light and medium-long wave infrared imaging. It briefly introduces the characteristics of indium gallium as a detector in shortwave infrared imaging systems, its development status at home and abroad, and the application of shortwave infrared imaging in different fields. Finally, future developments of shortwave infrared imaging are discussed.
The influence of the incident angle on the diffraction efficiency and microstructure height of the diffractive optical element (DOE) was analyzed to further study the influence of the incident angle and period width on the polychromatic integral diffraction efficiency (PIDE). Based on the extended scalar diffraction theory (ESDT), a mathematical model of the relationship among the microstructure height, incident angle, and period width of the DOE was established. An optimal design method for structural parameters, such as the design wavelength and microstructure height, was proposed based on maximizing the comprehensive PIDE (CPIDE) within a certain range of incident angles. A DOE operating within the infrared waveband was considered as an example. The results indicate that when the relative period width is 20 and the incidence angle range is 0° to 40°, the CPIDE of the DOE is 94.15%, and the microstructure height is 1.3396 ?m. This design method can realize the optimal design of a wide-angle DOE.
As the pixel size of the vanadium oxide (VOx) uncooled infrared focal plane decreased, the absorption area of the detector exhibited a sharp quadratic decrease in the edge length. Improving the absorption efficiency of the VOx uncooled infrared focal plane arrays has become a crucial research topic. In this study, a comprehensive and systematic simulation of the factors affecting the optical absorption of singlelayerand double-layer VOx uncooled detectors was conducted from the aspects of material and structure, in terms of the absorption characteristics of single-layer material, different absorption structures, height of the cavity, and thickness of the film system. A systematic approach to improve the absorption of VOx uncooled detectors is provided by quantitatively comparing the factors with the simulation results, which have certain reference significance for the design and research of VOx uncooled detectors.
For the registration task of infrared and visible binocular cameras with fixed relative positions, existing algorithms do not consider the prior fixed relative positions of the two cameras, resulting in problems, such as low registration accuracy, large differences in geometric positioning, and poor applicability. An infrared and visible binocular image registration method based on region search under geometric constraints. First, stereo correction was performed on the infrared and visible images using the calibration information of the infrared and visible binocular cameras, such that they were at the same height. Second, infrared and visible edge maps were obtained using phase congruency and feature points were extracted from the infrared edge map. Finally, a two-stage feature point search method is proposed to search for feature points with the same name in the local area of the visible edge map based on the infrared feature points. In the first stage, normalized cross-correlation (NCC) was used as a similarity metric to calculate the overall horizontal offset of the two edge maps, and the initial positions of feature points with the same name were predicted. In the second stage, a multiscale-weighted NCC was proposed as a similarity metric to accurately search for feature points with the same name around the initial location of feature points of the same name. Then, experiments were performed on the constructed real-environment dataset. The experimental results show that compared with other comparison methods, the number and accuracy of matching points and registration results in subjective vision are better.
This study introduces a multiscale sliding window filter (M-SWF) image fusion method to address issues with traditional filter banks in infrared and visible image fusion. First, a multiscale image decomposition method based on SWF is proposed to extract the structural detail layers and base layers of the source image. Second, the L1 norm fusion rule (L1-Fusion, L1F) is used to integrate the structural detail layers, which can extract the structure of the image. Then, to highlight the salient objects, energy attribute fusion (EAF), which is a rule for fusing image energy contributions, is used to integrate the base layers, and the fusion results are obtained by stacking the integrated multiscale structure detail layers and base layers. The energy contribution coefficient was analyzed, and a suitable energy contribution coefficient was obtained for the fusion of infrared and visible images in the M-SWF domain from subjective and objective perspectives. Compared with other fusion methods, the M-SWF not only improves the ability to extract the structural information of the source image but also improves the poor fusion effect and effectively highlights salient targets by integrating the energy attributes of the image.
To improve the quality of the fused image, the study presents a double-branch antagonism network (DANet) for the polarization direction images. The network includes three main modules: feature extraction, fusion, and transformation. First, the feature extraction module incorporates low and high-frequency branches, and the polarization direction images of 0°, 45°, 90°, and 135° are concatenated and imported to the lowfrequency branch to extract energy features. Two sets of polarization antagonism images (0°, 90°, 45°, and 135°) are subtracted and entered into the high-frequency branch to extract detailed features and energy. Detailed features are fused to feature maps. Finally, the feature maps were transformed into fused images. Experiment results show that the fusion images obtained by DANet make obvious progress in visual effects and evaluation metrics, compared with the composite intensity image I, polarization antagonistic image Sd, Sdd, Sh, and Sv, the average gradient, information entropy, spatial frequency, and mean gray value of the image are increased by at least 22.16%, 9.23%, 23.44% and 38.71%, respectively.
Infrared image segmentation plays a pivotal role in diagnosing faults in electrical equipment using infrared imagery. However, uneven heat dissipation, lower contrast, and interference from multiple sources of noise in electrical equipment can lead to over-segmentation of the target region, seriously affecting segmentation accuracy. In this study, we propose an Enhanced Fuzzy C-Means (EnFCM) clustering method based on fusion reconstruction for infrared image segmentation of electrical equipment. First, the gradientimage was subjected to an adaptive morphological reconstruction operation to ensure the segmentation ability of the algorithm on noisy images; second, the image was tested for saliency, and the reconstructed image was obtained by fusing the saliency map with the gradient map to highlight the features of the fault site and avoid over-segmentation; then, watershed segmentation was performed on the reconstructed image to obtain the super-pixel image; finally, the histogram clustering of the super-pixel image was obtained by segmentation. The experimental results on the infrared image of electrical equipment show that the algorithm in this paper can accurately segment the fault area on it, obtain its location and contour, and effectively improve the phenomenon of over-segmentation and in the comparison of the selected intersection and concatenation ratio and DICE coefficient indexes, this paper's method improves 81% and 79% on average compared to selected FRFCM, FCM, SFFCM, FCM~~SICM, RSSFCA, and AFCF; meanwhile, it is extremely robust to noise, and in the comparison of selected segmentation accuracy indexes, this paper's method achieves segmentation results that are on average 73% superior compared to selected FRFCM, FCM, SFFCM, FCM~~SICM, RSSFCA, and AFCF, thus, superior segmentation results were achieved.
The high heterogeneity of complex backgrounds destroys the low rank of a scene, and it is difficult for existing algorithms to use low-rank sparse recovery methods to separate dim targets from the background. To resolve this problem, this study transforms the dim target detection problem into a convex optimization function-solving problem for tensor models. It proposes a detection model based on sparsely enhanced reweighting and mask patch tensors. First, the stacked mask patch image was expanded into a tensor space, and a mask patch-tensor model was constructed to filter the candidate targets. Thus, a sparse enhanced reweighting model was constructed using structural tensors to suppress background clutter, and the limitation of setting the weighting parameters can be overcome by solving convex optimization functions. The experiments show that the proposed algorithm outperforms recent representative algorithms regarding the background suppression factor and signal-to-noise ratio gain, demonstrating its effectiveness.
Infrared and visible image fusion methods have problems such as insufficient information extraction, feature decoupling, and low interpretability. In order to fully extract and fuse the effective information of the source image, this paper proposes an infrared and visible image fusion method based on information bottleneck siamese autoencoder network (DIBF: Double Information Bottleneck Fusion). This method realizes the disentanglement of complementary features and redundant features by constructing an information bottleneck module on the twin branch. The expression process of complementary information corresponds to the feature fitting process of the first half of the information bottleneck. The compression process of redundant features corresponds to the feature compression process in the second half of the information bottleneck. This method cleverly expresses information extraction and fusion in image fusion as an information bottleneck trade-off problem, and achieves fusion by finding the optimal expression of information. In the information bottleneck module, the network obtains the information weight map of the feature through training, and uses the mean feature to compress the redundant features according to the information weight map. This method promotes the expression of complementary information through the loss function, and the two parts of compression and expression are balanced and optimized simultaneously. In this process, redundant information and complementary information are also decoupled. In the fusion stage, the information weight map is applied in the fusion rules, which improves the information richness of the fused images. Through subjective and objective experiments on the standard TNO dataset, compared with traditional and recent fusion methods, the results show that the method in this paper can effectively fuse useful information in infrared and visible images, and achieved good results on both visual perception and quantitative indicators.
The production process of coking enterprises generates abundant smoke. Their discharge and leakage can pollute the natural environment, endangering the safety of life and production. Considering the low contrast and poor texture of thermal imaging videos, this study detected smoke with motion and fuzzy characteristics. The noise degree of each frame image can be calculated to replace the fixed threshold of the Vibe detection algorithm so that the moving target area can be completely detected. First, the image was divided into block area images; then, the fuzzy-to-noise ratio in this area was extracted by combining the motion area, the features calculated when the fast fourier transform (FFT) was used to calculate the ambiguity were trained to generate a smoke classifier, and finally, the experimental video detection, with an average accuracy rate of 94.53%. The results show that the proposed algorithm is accurate, operates in real-time for smoke detection in infrared thermal imaging videos of coking enterprises, and has good anti-interference ability.
This paper proposes an infrared image fusion enhancement algorithm based on an improved wavelet threshold function and full-scale Retinex to address the problems of low signal-to-noise ratio, fuzzy detail, and poor clarity in existing infrared image enhancement algorithms. First, to overcome the degradation of infrared images caused by fixed-scale parameters and light scattering, a full-scale map of Retinex-scale parameters was obtained using atmospheric transmittance to improve image clarity. The input image and processed image with full-scale Retinex were used as the first and second inputs of the algorithm, respectively. Second, an improved wavelet threshold function was designed to solve the problems of artifacts and detail loss in the imagedenoising process of the traditional wavelet threshold function. The threshold function introduces a scaling factor that can be adjusted adaptively according to the number of layers after calculating the wavelet coefficient of the high-frequency subgraph of each layer. An adjustment factor was introduced and combined with an exponential function to suppress the high-frequency subgraph noise and preserve detailed information. The high- and low-frequency subgraphs of the above two inputs were then fused using wavelet image fusion to improve the texture details of the output images. The simulation results demonstrate that the proposed algorithm outperforms other comparison algorithms regarding noise reduction and detail highlighting capabilities, enhancing the visual quality of infrared images for the human eye. Finally, this algorithm was applied to enhance infrared images collected by an infrared imaging module, and the experimental results showed that the proposed method is practical.
This study designed a buried channel metal-oxide-semiconductor (BCMOS) image sensor experiment to address increased dark current caused by low-energy electron bombardment (300 eV to 1500 eV) on the alumina passivation layer. For a CMOS image sensor with a 10 nm alumina passivation layer, an increase in the dark current rate is obvious when the bombardment energy is greater than 600 eV. When the bombardment electron energy does not exceed 1.5 keV, the dark current has a maximum value of about 12000 e-/pixel/s. Finally, after electron bombardment, the dark current of the CMOS image sensor decreased exponentially when the sensor was placed in an electronic drying cabinet. The main reason for the above phenomenon is the increased defect states at the interface between alumina passivation layer and silicon caused by incident electrons.
When operating mechanical equipment, the number of fault samples marked is small, which leads to low accuracy of the fault diagnosis of the established model. Therefore, this study proposes a defect detection method for eddy current thermal imaging of a workpiece that combines depth learning and domain adaptation. First, the attention mechanism is introduced into the deep residual network ResNet50 to enhance the feature extraction capability of the model. Then, the source and target domain data are sent into the improved ResNet50 network to extract the depth features. The local maximum mean difference is introduced into the full connection layer of the network to reduce the distribution difference between the two domain features to achieve the distribution alignment of related sub-domains. Finally, workpiece metal material defects were detected in the Softmax classifier of the network. The experiment was conducted on the open magnetic tile dataset and eddy current infrared image dataset of the metal plate collected during the experiment. The results show that the method proposed in this paper is highly accurate in detecting and recognizing crack defects in eddy current infrared images. The advantages of the method in this study were verified by visualizing the analysis results using the t-distribution random neighbor embedding method.
The spontaneous combustion of residual coal in a goaf is a major cause of underground fire accidents in coal mines. Owing to the actual situation of underground goafs, existing detection methods cannot directly determine the true situation of residual coal thermal storage and spontaneous combustion. This study utilized the non-contact temperature measurement characteristics of the infrared thermal imager, and through the analysis of temperature measurement principles, experimental system design, and construction methods, conducts an experiment to measure the effective detection distance of the FOTRIC348 infrared thermal imager close to the real environment of the underground goaf, and verifies it in underground engineering. The results show that the effective detection distances of the FOTRIC348 in the regional, line, and point temperature measurement modes were 8 m to 12 m, 10 m to 13 m, and 9 m to 13 m, respectively. The effective detection distance of FOTRIC348 during underground testing was 10 m to 12 m. Based on the intersection of the above detection distances, the effective detection distance of the FOTRIC348 infrared thermal imager for determining the thermal storage and spontaneous combustion stage of residual coal in the goaf of the two parallel grooves of the working face is 10 m to 12 m. Determining this value can provide a targeted decisionmaking basis for implementing prevention and control measures for spontaneous coal combustion in the goaf.