Advanced Materials Research Vol. 981

Abstract: This paper proposes a unified architecture for computation of discrete cosine transform (DCT) and its inverse transform (IDCT). The matrix decomposition algorithm is used to deduce the proposed algorithm. Based on this algorithm, a unified DCT/IDCT architecture is developed. Then, this architecture is modeled in HDL, verified and implemented with FPGA. Experiment results show that the unified DCT/IDCT architecture has low hardware complexity and high calculation accuracy.

Abstract: In this paper, a new color image encryption mechanism based on multiple chaotic systems is proposed. In the proposal, two modules are achieved by mixing the features of horizontally and vertically adjacent pixels with the help of adopted multiple chaotic systems, respectively. Then, substitution/confusion is accomplished by generating an intermediate chaotic key stream image based on the adopted chaotic maps. Experimental results show that the proposed scheme is able to manage the trade-offs between the speed performance and security requirements.

Abstract: In this paper, we introduce the high performance Deformable part models from object detection into human action recognition and localization and propose a unified method to detect action in video sequences. The Deformable part models have attracted intensive attention in the field of object detection. We generalize the approach from 2D still images to 3D spatiotemporal volumes. The human actions are described by 3D histograms of oriented gradients based features. Different poses are presented by mixture of models on different resolutions. The model autonomously selects the most discriminative 3D parts and learns their anchor positions related to the root. Empirical results on several video datasets prove the efficacy of our proposed method on both action recognition and localization.

Abstract: For the most commonly used method of are susceptible to noise, it is difficult to extract a complete moving target . Aiming at this problem, we propose a video segmentation algorithm through the combination of an improved Kirsch edge operator and the three-frame difference. Firstly, we got a moving region through a differential by the three consecutive image frames, and then got the edge on the current frame using an improved Kirsch edge detection operator, Synthesis of the two test results, we got a more accurate moving object edges. Then using the edge connector algorithm to complete the connection of the fracture edge, and finally, got moving target mask image through regional filling, thereby dividing the complete moving target, thus dividing the complete moving target. Experimental results show that the algorithm has a certain robustness, and can accurately detect moving targets.

Abstract: The incidental component in addition to the measured target signals is considered as noise of Positron Emission Tomography (PET) images. A novel method to denoise the PET images based on Empirical Mode Decomposition (EMD) and Independent Component Analysis (ICA) associated with Sparse Code Shrinkage (SCS) technique is proposed in this paper. EMD is executed to decompose a PET image into a number of Intrinsic Mode Functions (IMFs), which are used to reconstruct a new PET image after chosen by means of an inverse EMD procedure. By applying ICA to the new PET image, an orthogonal dataset can be obtained and the signal-noise separation can be realized. Then a clearer PET image can be reconstructed by SCS. The simulation results indicate that the proposed method is effective to denoise PET images.

Abstract: Image segmentation is an important part of the image process, and it is also the current hot and focus in image research. How to achieve better segmentation results are dominating targets of researchers. Currently, image segmentation based on clustering is the main research area. Firstly, this paper introduces the traditional C-means clustering algorithm and its characteristic has been analyzed. Then, the initial clustering center and the number are selected using the histogram. Finally, the image is converted from the RGB space to Lab space for clustering, and it has improved the accuracy and efficiency of image segmentation.

Abstract: In coded-structured light three dimensional system, system calibration plays a vital role for the measurement accuracy. The camera calibration method is very mature, but the study about projector calibration is less. Therefore, this paper proposes a projector calibration method with simple calibration process and high accuracy. This method combines the Zhang’s plane model calibration method with orthogonal phase shift coding. In calibration process, this paper uses phase shift coding pattern to establish the relationship of projector image and camera corner point coordinates. According to the image coordinates in the projector’s perspective, we program and calculate the projector’s internal and external parameters matrix based on the Zhang’s plane model calibration toolbox. The results show that the proposed method is simple and flexible, the maximum relative error of the calibration parameters is 0.03%, and it meets the requirements of system calibration in medical or industrial fields.

Abstract: A stereo matching algorithm was proposed based on pyramid algorithm and dynamic programming. High and low resolution images was computed by pyramid algorithm, and then candidate control points were stroke on low-resolution image, and final control points were stroke on the high-resolution images. Finally, final control points were used in directing stereo matching based on dynamic programming. Since the striking of candidate control points on low-resolution image, the time is greatly reduced. Experiments show that the proposed method has a high matching precision.

Abstract: A high-precision calibration method was proposed. This method is divided four steps: extracting calibration data, building model, calculating inside and outside parameters, and correcting camera distortion. Experimental results show that calibration is very accurate and the total error is not more than 0.06 pixels.

Abstract: Image content based retrieval is an important research area with application to digital libraries and multimedia databases. Its image characterization and similarity measure must closely follow perceptual characteristics. In this work, a new image retrieval method is proposed by combining color features and texture features based on Dempster-Shafer (D-S) theory. In this proposed method, Multi-color features included color histogram, color moment and color correlogram are used for color analysis, and gabor wavelet features is employed for texture analysis. Then these features are modeled as a mass function in evidence theory, and color and texture detection results are fused at decision-making level. The experimental results show that the proposed method takes advantage of the respective merits of color and texture features and therefore improves retrieval accuracy and reduces recognition error rate.