In hyperspectral image (HSI) analysis, one of the most important tasks is targetdetection, requiring the execution of algorithms with high computational complexity. Recently, research efforts have focused on on-board...
详细信息
In hyperspectral image (HSI) analysis, one of the most important tasks is targetdetection, requiring the execution of algorithms with high computational complexity. Recently, research efforts have focused on on-board real-time targetdetection to provide timely responses for swift decisions. Therefore, it is necessary to use a technology that provides the performance needed for real-time targetdetection, and at the same time meets the satellite payload requirements. Field-programmable gate arrays (FPGAs) have very interesting properties in terms of performance, size, and power consumption, which have become the standard option for on-board processing. In this letter, we present a hardware optimized implementation for FPGAs of the automatic targetdetection and classification algorithm (ATDCA) using the GramSchmidt (GS) method for orthogonalization purposes. The ATDCA-GS algorithm is directly coded using VHDL and verified on a Virtex-7 XC7VX690T FPGA using real hyperspectral data [collected by Hyperspectral Digital Imagery Collection Experiment (HYDICE) sensor and by NASAs Airborne Visible/infrared Imaging Spectrometer (AVIRIS)] and a synthetic image. Experimental results demonstrate that our hardware version of the ATDCA-GS algorithm outperforms previous implementations (multicore processors, GPUs, and accelerators) in both computation time (obtaining real-time performance) and power consumption, demonstrating the suitability of FPGAs for this purpose.
A knowledge-based approach to the detection, tracking, and classification of ground-based formations of point targets in sequences of digitized forward-looking infrared (FLIR) image sequences is presented. The system ...
详细信息
A knowledge-based approach to the detection, tracking, and classification of ground-based formations of point targets in sequences of digitized forward-looking infrared (FLIR) image sequences is presented. The system has two components: a point target detector and tracker (PTD) which processes the image sequences and supplies candidate point targets to the knowledge-based target formation detector for clustering into formations, which are then classified as linear, V, and the like. The system has been implemented in software written in C and Common Lisp and evaluated on a variety of FLIR image sequences. The results indicate that, in all cases, the knowledge base improves system performance over that attained by the PTD alone.< >
The classification of small, low-observable airborne targets, such as drones and birds, poses significant challenges due to their low detection rates. Conventional vision sensorbased approaches often suffer from reduc...
详细信息
ISBN:
(数字)9798331506940
ISBN:
(纸本)9798331506957
The classification of small, low-observable airborne targets, such as drones and birds, poses significant challenges due to their low detection rates. Conventional vision sensorbased approaches often suffer from reduced performance in low-visibility environments or adverse weather conditions. Additionally, the integration of infrared sensors alongside camera sensors increases hardware complexity and cost, rendering such solutions inefficient. To address these limitations, we propose a method that leverages deep learning and inverse synthetic aperture radar (ISAR) imaging for accurate targetclassification, using only radar sensors. Our proposed deep learning-based ISAR image classifier comprises two key components: simulated ISAR image generation and deep learning-based classification. We construct simulated ISAR datasets using point scatter (PS) modeling for quadcopter drones, hexacopter drones, and aircraft, and three-dimension (3D) mesh modeling for birds, unmanned aerial vehicles, and quadcopter drones. The two datasets based on PS and 3D mesh modeling are used to train a proposed deep learning classifier. The proposed classifier can achieve a classification accuracy of $\mathbf{9 8 \%}$ on the PS-based dataset and $\mathbf{9 6 \%}$ on the 3D mesh-based dataset, where scattering was calculated using the physical optics method.
In recent years, RADARs have begun to replace more sensors like cameras, ultrasonic devices, and infrared sensors used in common places in everyday life. As integrated circuit technology has advanced, the environmenta...
详细信息
In recent years, RADARs have begun to replace more sensors like cameras, ultrasonic devices, and infrared sensors used in common places in everyday life. As integrated circuit technology has advanced, the environmental robustness of radar technology has enabled small, inexpensive, short-range radars operating at frequencies ranging from a few GHz to hundreds of GHz (millimeter waves). As a result, RADAR has become a popular choice for many new applications, including patient monitoring, adaptive control of self-driving cars and drones, structure monitoring, airport security, and gesture recognition. Another recent area of interest is RADAR-based wildlife anti- poaching activities. The model's performance is assessed using the spectrogram obtained from a radar signature database of moving targets in a typical setting. The goal of this work is to discriminate between moving human and animal targets based on variation in range of red cells in the spectrogram. Observed results indicate that the model's classification accuracy depends on the length of continuous on target observation. The radar-based targetdetection and classification system used for border security measures and wildlife anti-poaching operations for animals like rhinos or elephants are both potential activities.
In order to improve the infrareddetection and discrimination ability of the smart munition to the dynamic armor target under the complex background,the multi-line array infrareddetection system is established based ...
详细信息
In order to improve the infrareddetection and discrimination ability of the smart munition to the dynamic armor target under the complex background,the multi-line array infrareddetection system is established based on the combination of the single unit infrared *** surface dimension features of ground armored targets are identified by size calculating solution *** signal response value and the value of size calculating are identified by the method of fuzzy recognition to make the fuzzy classification judgment for armored *** to the characteristics of the target signal,a custom threshold de-noising function is proposed to solve the problem of signal *** multi-line array infrareddetection can complete the scanning detection in a large area in a short time with the characteristics of smart munition in the steady-state scanning *** method solves the disadvantages of wide scanning interval and low detection probability of single unit infrared *** reducing the scanning interval,the number of random rendezvous in the infrared feature area of the upper surface is increased,the accuracy of the size calculating is *** experiments results show that in the fuzzy recognition method,the size calculating is introduced as the feature operator,which can improve the recognition ability of the ground armor target with different shape size.
infrared point targetdetection based on markov random field (MRF) is mainly formulated as a binary classification problem, leading to a poor adaptability to complex background with high false alarm rate. This paper f...
详细信息
infrared point targetdetection based on markov random field (MRF) is mainly formulated as a binary classification problem, leading to a poor adaptability to complex background with high false alarm rate. This paper formulates the infrared point targetdetection as a multi-classification problem, and proposes a detection method based on multi-label generative MRF (MG-MRF) model. First, the MG-MRF model is proposed and the optimal label configuration of the infrared image is derived using iterated condition mode (ICM). Second, the pointwise adaptive filter is structured utilizing local labels to suppress the background clutter. Finally, an adaptive threshold is utilized to segment the target in the residual image. The experimental results on various backgrounds demonstrate that the detection method based on MG-MRF has a strong suppression of false alarm with superior performance in terms of accuracy and efficiency. (C) 2017 Elsevier B.V. All rights reserved.
暂无评论