Applied Mechanics and Materials Vols. 333-335

Abstract: This paper proposes a fast GrowCut (FGC) algorithm and applies the new algorithm in three-dimensional (3D)kidney segmentation from computed tomography (CT) volume data. Users could mark the object of interest with different labels in CT slices.FGC propagates the labels using monotonically decreasing function and color features to derive an optimal cut for a given data in space. The color features play a great role in comparing with neighborhood cells. The experimental results clearly demonstrate the superiority of FGC in accuracy and speed.

Abstract: The process and practical results of the image processing technology of monocular vision aided target measurement are given. Firstly, the image was enhanced by mathematical morphology and segmented wih geometry-based feature extraction; Then according to least square ellipse fitting algorithm,the center coordinates of mark point was accurately striked; Finally, improved algorithm based on the elliptical error evaluation was presented, which had high accuracy and good stability. The experiments show that the shape error is about one pixel, and it is reduced significantly after the algorithm improvement.This method can be widely used in monocular vision measurement.

Abstract: This paper discusses land utilization and classification based on remote sensing image analysis expert system technology. Taking the bureau department area remote sensing images as the basic information, and using MAPGIS and Ecognition software, we carry on the classification of the land. The results show that remote sensing image analysis expert system technology applies to land utilization situation and land classification, which has reference value.

Abstract: In view of the situation that most image fusion methods make spectral distortion more or less; this paper proposes a spatial projection method by introducing the Gaussian scale space theory. According to mechanism of the human visual system represented by the Gaussian scale space, the spatial details feature information are extracted from the original panchromatic (PAN) and multispectral (MS) images, then the feature differences are projected into the original MS image to obtain the fused image. The experimental results with WorldView-2 images show that the proposed method can improve the spatial resolution of the fused MS image effectively, while it can make little spectral distortion to the fused images so as to maintain the great majority spectral information.

Abstract: As essential character of urban region, building extraction and recognition has been applied broadly in urban mapping, urban planning and population census. Traditional manual plotting is time consuming and expensive, which therefore challenges for automatic or semi-automatic solutions. High-resolution multi-spectral remote sensing imagery provides both spectral and spatial information for acquiring urban features to update geographic information database. An advanced algorithm based on the combined use of spectral and spatial features will be developed and employed to recognize and extract buildings from multi-spectral imagery in this paper. Firstly, the imagery is spatially filtered to achieve more homogeneous regions. With the spectral and spatial features, an automatic and iterative region growing algorithm is employed to segment the imagery. A feature vector is developed to recognize the buildings from the final segmentation result. The result shows that this method can extract 69.8% of the buildings in the tested imagery.

Abstract: Drawing on the suvey of visual attention degree and its significance in psychology and physiology , in recent years, researchers have proposed a lot of visual attention model and algorithms, such as Itti model and many saliency detection algorithms. And in recent years, the researchers applied the visual attention of this technology in a lot of directions, such as a significant regional shifts and visual tracing detection model based on network loss, for video quality evaluation. This paper summarizes the various algorithms and its application of visual attention and its significance.

Abstract: The azimuth resolution of imaging sonar lies on the aperture of the sensor array. In the paper, the Multiple-input Multiple-output Synthetic Aperture Sonar (MIMO-SAS) composite acoustic imaging technique was applied to imaging sonar equipped on the movable platform. The azimuth resolutions of both cross-track direction and along-track direction were obtained. The simulation results demonstrate that comparing with the conventional imaging technique Single-Input Multiple-output (SIMO), the azimuth resolution of cross-track direction is improved since the virtual array aperture of MIMO technique, and the azimuth resolution of along-track direction is obtained since the synthetic aperture of SAS technique. The simulation results are consistent with the theoretic value of the azimuth resolution.

Abstract: In order to effectively improve the imaging quality of the nighttime image, An improved wavelet threshold denoising algorithms which could make up the shortage of soft and hard threshold function was researched in this paper. The denoising effect is much better when the parameter e is adjusted to a certain value. The simulation result shows that the gaussian noise in the night image is removed effectively. The Pseudo-Gibbs, Ringing effects, the lack of details and edge blur were overcome obviously.

Abstract: In this paper, we present a blind digital watermarking method for color image based on nearest neighbor interpolation. The algorithm has the following characteristics: (1) it generate adaptive watermark based on the size of carrier image,so as to improve the robustness of binary watermark; (2) the watermark is embedded in the DCT coefficient of intermediate frequency, which can effectively resist common attacks, such as Gaussian noise-adding, salt and pepper noise-adding, cropping, Wiener filtering, JPEG compression;(3) the watermarking detection is simple and fast, and does not require carrier image. The experiments show that the algorithm has good invisibility and strong robustness.

Abstract: The recognition and localization of cooperative objects are very important for spacecraft pose estimation towards autonomous rendezvous and docking (RVD). In this paper, an adaptive threshold segmentation algorithm is proposed base on weighted maximum gray value for multi-object detection, and eight-neighborhood region growing is employed for multi-object recognition. In order to achieve high-accurate localization, a sub-pixel object positioning approach is put forward by combination bilinear interpolation with median filtering, which employs bilinear interpolation to calculate sub-pixel centroid for reducing algorithm systematic errors, and applies median filter to reduce random errors produced by image noises. The experimental results show that the proposed algorithms are feasible and effective with high positioning accuracy of 0.01 pixels, and have outstanding advantages of anti-disturbance and real-time performance, thus can satisfy the practical requirements in the visual measurement and pose estimation of cooperative objects for the RVD in space exploration.