Applied Mechanics and Materials Vols. 347-350

Abstract: The blind have so many inconveniences in their lives. In recent years, more and more blind sidewalks are designed and built for them. This is good news for the blind. However, people would consciously or unconsciously leave some stuff on the blind sidewalk. Therefore, the occupation of blind sidewalk is a serious problem gradually. We decided to develop an application on smartphone to help the blind to avoid troubles referred above. Within the camera on the smartphone, the application detects obstacles on the blind sidewalk based on image processing and warns the blind that there are obstacles in front of him. This paper proposed an approach on detecting the direction of blind sidewalk after the blind leave the blind sidewalk to avoid the obstacles. We used several algorithms of image processing and conduct several experiments to exam this system.

Abstract: For uninhabited flying vehicle, it is a key prerequisite of truly autonomous mobile vehicles to simultaneously localize and accurately map its surroundings. Kalman filter-based algorithms require time quadratic in the number of landmarks to incorporate each sensor observation. This paper presents an algorithm so called FastSLAM that recursively estimates the full posterior distribution over robot pose and landmark locations, but scales logarithmically with the number of landmarks in the map. FastSLAM factors the posterior into a product of conditional landmark distributions and a distribution over UAV paths. The algorithm has been tested in UAV environments. Experimental results demonstrate the advantages and disadvantages of the FastSLAM algorithm for UAV.

Abstract: Human detection is a challenging problem, owing to variations in pose, body shape, appearance, clothing, illumination, and background clutter, in addition, the cameras or backgrounds make it even harder. But even so, it has many potential applications including net-meeting, security, human-computer interaction, gaming, and even health-care. Various new approaches have been proposed to solve this problem. We have studied and implemented a method by using dynamic contour [ and Histograms of Oriented Gradients [ to detecting human body fast and accurately in static images.

Abstract: Non-line-of-sight (NLOS) is one of the main factors that affect the ranging accuracy in wireless localization. This paper proposes a two-step optimizing algorithm for TOA real-time tracking in NLOS environment. Step one, use weighted least-squares (WLS) algorithm, combined with the NLOS identification informations, to mitigate NLOS bias. Step two, utilize Kalman filtering to optimize the localization results. Simulation results show that the proposed two-step algorithm can obtain better localization accuracy, especially when there are serious NLOS obstructions.

Abstract: A military vehicle hoisting jack does not work well as a result of its brackets frequent serious offset deformation during use. To solve this problem, the finite element method is applied to analyze the structure of the bracket. In succession, the dimension, chamfer and material are improved with respect to the analysis results. The problem mentioned above has disappeared for more than two years since the improved bracket put into use. The proposed method is valuable.

Abstract: This paper studies the real-time compression and storage technology,extraction and collection contains that Images,Image compression standard,Image storage.The paper use JPEG2000 compression standard.Image via IDE interface to hard disk storage.The results show that non-interest areaimage compression and srorage can remove redundant;reduce the amount of data transfer and increase the transmission speed.

Abstract: This paper briefly introduced the development of video face detection and point out the shortage of current face detection system that may produce much of false alarms. Then we detail the classic Viola face detector which using integral image, Haar-like features and AdaBoost algorithm for training. Compared with Viola face detector, we proposed an available multi-model fusion method to reduce false alarms in video face detection that is combining head-shoulder detector with HOG features. After introduced the related knowledge of HOG features, we proposed a fusion detector structure which can improve the accuracy and efficiency of detection.

Abstract: In this paper, firstly we try to look for ways to avoid the camera parameters in order to reconstruct 3D model. We attempt to use the parallel stereo visual system and carry out the mathematical derivation of argumentation. Then we use epipolar geometry to solve this problem. And compare the computation algorithms of fundamental matrix. Then for the algorithm, we propose some improvement to compute the fundament matrix more precisely so that the algorithm is more stable and the robustness is stronger.

Abstract: Face recognition continues to be a hot topic in pattern recognition and computer vision recent years. Meantime, sparse representation is a new technique utilizing compressed sensing method applied in pattern recognition research recent years. This paper comparison with PCA and SP for face recognition, makes some improvement on great performance eigenface algorithm, using sparse representation with eigenface space to achieved better result. Finally, the proposed algorithm based on eigenface sparse recognition shows that of better result.

Abstract: This paper carries out an algorithm research on bridge target detection in SAR images and presents a method that combines both texture features and correlation features. The method firstly extracts initial targets by using the algorithm of histogram equalization segmentation, and then conducts a contrastive analysis for targets and their surrounding background textures by using the gray level co-occurrence matrix to get rid of the false alarm target. The experimental results show that the method is simple, effective and has certain algorithm robustness.