A5017172632- Advanced Measurement and Detection Methods
- Infrared Target Detection Methodologies
- Optical Systems and Laser Technology
- Remote-Sensing Image Classification
- Ocular and Laser Science Research
- Remote Sensing and Land Use
- Advanced Optical Sensing Technologies
- Advanced Image Fusion Techniques
- Laser Design and Applications
- Image Processing Techniques and Applications
- Chaos-based Image/Signal Encryption
- Optical Polarization and Ellipsometry
- Semiconductor Lasers and Optical Devices
- Advanced Steganography and Watermarking Techniques
- Advanced Measurement and Metrology Techniques
- Radiation Therapy and Dosimetry
- Image and Object Detection Techniques
- Advanced SAR Imaging Techniques
- Optical measurement and interference techniques
- Nuclear Physics and Applications
- Education and Work Dynamics
- Integrated Circuits and Semiconductor Failure Analysis
- Remote Sensing in Agriculture
- Cryptographic Implementations and Security
- Engineering Education and Curriculum Development
PLA Army Engineering University
2018-2024
Shijiazhuang University
2018-2024
Ordnance Engineering College
2007-2017
Beijing Institute of Optoelectronic Technology
2009
Peking University
2005
Due to the limitations of spatial resolution and detector level, traditional hyperspectral image target detection focuses more on spectral analysis, morphology information is not fully utilized in detection. The Constrained Energy Minimization (CEM) method a classic algorithm that can highlight target, suppress background information, achieve effect separating from image. However, CEM supervised requires obtaining advance. various factors such as material composition, object shape, imaging...
Hyperspectral imaging can simultaneously obtain the spatial morphological information of ground objects and fine spectral each pixel. Through quantitative analysis characteristics objects, it complete task classification recognition objects. The appearance spectrum technology provides great advantages for military target detection promotes continuous improvement reconnaissance levels. At same time, camouflage materials methods that are relatively resistant to hyperspectral also developing...
Abstract Target detection based on hyperspectral images refers to the integrated use of spatial information and spectral accomplish task localization identification targets. There are two main methods for target detection: supervised unsupervised methods. Supervision method differences between be tested surrounding background identify when spectrum is known. In ideal situations, object algorithms perform better than algorithms. However, current mainly have problems: firstly, impact...
Hyperspectral reconnaissance technology can realize three-dimensional by using target space and spectral information, which effectively improves the efficiency of battlefield reconnaissance. However, in order to obscure what is true false confuse enemy, camouflage also developing. Hiding background environment setting targets have become common procedures on battlefield. The camouflaged has very similar spatial characteristics real target, so method identifying according similarity threshold...
Hyperspectral image target recognition, as an important component of computer vision, has broad application prospects in the military field. We conducted research on hyperspectral recognition technology based deep learning, with typical targets main objects. A algorithm suitable for images is proposed YOLO network model. To address issue not being sensitive enough to small targets, calculation formula loss function been adjusted, and a multi-scale training method added. In response...
Neutron detection efficiency for a long wedge-shaped plastic scintillation detector was measured by using the Am/Be neutron source. The overall attains 23% at around 2 MeV and decreases to 15% about 5 MeV. experimental results are in agreement with Monte Carlo simulation calculations.
In imaging laser warning system, the spatial distribution of CMOS light-sensitive sensor array is expounded. When locating center spots, non-light-sensitive area makes distortion spot point spreading function, so it induces error. Fill factors with different value affect precision are discussed, and diameter spots affected by fixed fill factor analyzed. The standard deviation emulated Monte Carlo analysis, an experiment made to verify analysis. experimental result consistent It has guiding...
With the extensive use of laser equipment and weapons systems, such as range finder guided weapons, energy each source is greatly different. So requirement higher dynamic presented in detection system. For problem insufficient Imaging Laser Detection System (ILDS), a method by image processing to extend system discussed based on existing hardware wide-angle telecentric fish-eye lens CMOS camera device. Frame subtraction technique commonly used ILDS recognize locate spot filtering out...
The demand for eye-safe 1550 nm pulsed semiconductor laser-emission modules is increasing in the field of active laser detection, owing to their long range and high precision. power narrow pulse these can significantly improve distance accuracy active-laser detection. Here, we propose an equivalent circuit model a multiquantum-well based on structure device. We developed design method according InGaAlAs laser. In this method, module was divided into chip laser-driver levels optimization...
In image-forming laser locating system, irregular diffuse spot is usually generated by the factors as follow: orientation, intension and multi-path of beam. A precise algorithm for based on Gradient Hough Transform presented in this paper. First all, gradient gray image produced improved Sobel operators. Then segmented certain threshold. And then center localized an transform algorithm. The experiment shows that invalid accumulation Randomized (RTH) effects orientation are well resolved,...
Dynamic infrared camouflage can need to change according the background, and quickly from a state of into another camouflage. The image characteristics various states weak correlation, make surveillance tracking, target real difficulty guidance system, unable complete reconnaissance combat, so as improve battlefield survivability target. According different background conditions, algorithm is adopted realize for equipment, that equipment be integrated. evaluation effect needs carried out...
As one of the main weapons, impulse laser rangefinders have become object electro-optical countermeasures. So its real maximum range (defined as utmost operating in paper) becomes most concerned index to evaluate performance countermeasure weapons. A method for calculating rangefinders′ by sensitivity different weather is obtained. Then a experiment getting supplied. By analyzing data which detectivity 40%-60%, 1.7×10<sup>-5</sup> W 9.8×10<sup>-5</sup> W. For reason that order...
The imaging spectrometer is limited by short response time and narrow channel, resulting in low signal-to-noise ratio of hyperspectral images. accurate estimation noise has a significant impact on some preprocessing downstream tasks. existing methods for images are all focused satellite aviation data, there little research with high spatial resolution. For this type image, This article proposes method based stacked autoencoder. Firstly, the image divided into multiple uniform regions using...
Hyperspectral image refers to a three-dimensional data cube containing both spatial shape information and spectral reflectance of ground objects. In recent years, with the rapid development land-based platforms near flight carriers, production application hyperspectral imaging equipment are gradually mature. Compared images under remote sensing conditions, conditions have higher resolution temporal resolution, saving lot time use costs. The technology provides solution problem "similar is...
Abstract Spectral uncertainty is one of the most prominent spectral characteristics hyperspectral images. Compared to process remote sensing imaging, imaging under land-based conditions has variable detection directions, random times, and complex environmental conditions, resulting in increased targets The target mainly refers phenomenon “Same reflectance but different objects” object reflectance” which poses significant difficulties for subsequent image recognition. In order analyze images...
Air target recognition in real-world scenarios has become an important part of the military offensive and defensive systems various countries. By identifying aerial targets captured by image acquisition equipment utilizing obtained information to achieve effective identification friend or foe, enemy sources, combat capabilities, intentions, references are provided for tactical decision-making. With continuous development technology, traditional manual based methods no longer capable targets....