Derong Chen

Learn More
We propose a new integrated target tracking, recognition and segmentation algorithm, called ATR-Seg, for infrared imagery. ATR-Seg is formulated in a probabilistic shape-aware level set framework that incorporates a joint view-identity manifold (JVIM) for target shape modeling. As a shape generative model, JVIM features a unified manifold structure in the(More)
A new joint view-identity manifold (JVIM) is proposed for multiview shape modeling that is applied to automated target tracking and recognition (ATR). This work improves our recent work where the view and identity manifolds are assumed to be independent for multi-view multi-target modeling. A local linear Gaussian process latent variable model (LL-GPLVM) is(More)
We propose a new joint view-identity manifold (JVIM) for multi-view and multitarget shape modeling that is well-suited for automated target tracking and recognition (ATR) in infrared imagery. As a shape generative model, JVIM features a novel manifold structure that imposes a conditional dependency between the two shape-related factors, view and identity,(More)
This paper constructs a set partition coding system (SPACS) to combine the advantages of different types of set partition coding algorithms. General tree (GT) is an important conception introduced in this paper, which can represent tree set and square set simultaneously. With the help of GT, SPIHT is generalized to construct degree-k SPIHT based on the(More)
A new probabilistic model called ATR-Seg for automated target tracking, recognition and segmentation is proposed that incorporates a shape constrained level set with a shape generative model along with motion model. The shape model involves a view-independent identity manifold and infinite identity-dependent view manifolds for multi-view and multi-target(More)
In order to improve the efficiency of algorithm, parameter optimization for hyperspectral image compression algorithm of maximum error controllable has been studied in this paper. Firstly, a mathematic optimal model for the hyperspectral image compression ratio was established. Secondly, we analyzed the model and simplified it by Gaussian function. Finally,(More)
A method of target recognition based on edge features of the images acquired before and after the damage is proposed to improve the accuracy of registration and recognition in Battlefield Damage Assessment. The image registration combined circle area method and angular point extraction is conducted after the preprocessing. The vector groups of edge features(More)
This paper presents an experimental investigation of the performance of an orthogonal frequency division multiplexing (OFDM) system in slow fading channels. The system has an 80 dB dynamic range and operates at 340 MHz with an 8 MHz 3 dB RF bandwidth. The peak user data rate achieved is 24 Mbits/sec. A novel automatic gain control (AGC) algorithm is(More)
A weighted pose estimation method using lines that intersect at a point is proposed. Because of the weak constraints, the accuracy of pose estimation using lines in this configuration is sensitive to the error of a two-dimensional (2D) line parameter. Therefore, we construct the objective function directly based on weighted image line points instead of 2D(More)
In this paper, two kinds of optimization techniques were proposed to enhance the performance of H.264 decoder that based on a DSP processor. One approach is to transfer reference data via DMA with a novel approach, which can utilize external bus efficiently. The other approach is to perform the block-based de-blocking filter, which minimizes the times of(More)
  • 1