By applying compressive sensing to infrared imaging systems, the sampling and transmitting time can be remarkably reduced. Therefore, in order meet real-time requirements of small target detection tasks remote field, many approaches based on have been proposed. However, these need reconstruct image from domain before detecting targets, which is inefficient due complex recovery algorithms. To overcome this drawback, paper, we propose a two-dimensional adaptive threshold algorithm for...

This work presents a spectral matching algorithm of missile plume detection that based on neural network. The radiation value the characteristic spectrum tail flame is taken as input network's structure including number nodes and layers determined according to bands types. We can get network weight matrixes threshold vectors through training using samples, we determine performance testing test samples. A small amount data cause has advantages simple practicality. Network composed matrix...

PID control algorithm has the advantages of precision and stability, but it is difficult to meet requirements complex control, improved in urgent need development.Fuzzy combined with fuzzy PID, characteristics fast stable, its application more extensive.Controller a kind encapsulation data, method interface technology, this paper use VC6.0 design controller algorithm, "zutaiwang" running environment double let water tank liquid level experiment shows that compared faster achieve both are no...

Aimed at space-based passive detection and tracking of ballistic targets, a multi-target multi-Bernoulli (MeMber) filtering algorithm based on focal plane ambiguous measurement model is proposed. The error sources are analyzed. It found that target the basis space in framework distributed data processing structure, main pixel resolution. Based above analysis, established to replace traditional generalized likelihood function designed. Finally, MeMber filter modified function. simulation...

Aiming at the problem of multi-target tracking accuracy decreasing and target SNR unknown in clutter environment, a algorithm based on grayscale feature is proposed. In this algorithm, likelihood function estimation prior information are constructed by using historical measurement information. When updates state, introduced into original function, used to increase discrimination between clutter, so as filter out more improve accuracy. Simulation results show that can achieve better effect...

To reduce the cardinality bias in multi-target tracking, a novel multi-Bernoulli filter called spectral matching based generalized labeled is derived. By introducing information of target update step, proposed can improve tracking performance. Under same assumption as GLMB and CPHD recursion, recursion unbiased estimation. In addition, sequential Monte Carlo implementation proposed. Simulation results show that has better performance state number estimation than filter.

Based on the non-local similarity of image, low rank block image is obtained by reconstruction, which basic method to apply robust principal component analysis (RPCA) algorithm single frame infrared small target detection. This paper introduces application process RPCA in detection and analyzes influence various blocking methods under different backgrounds. In order solve problem that it difficult select window sliding step segmentation complex background, a selection based larger value...