Advanced Materials Research Vols. 317-319

Abstract: Extraction of feather defects is difficult because of their complicated and diversified characteristics. It is proposed that feather stain defects be extracted in HSI color space according to color characters of different types of feather defects. The improved variational level set method is adopted to locate and segment the feather stains on that basis. Experiment results show that the proposed methods can extract feather stains effectively and segment them accurately.

Abstract: Through developing intelligent system of human-like thinking, national defense, economic, education, culture and so on will be impact. Constructing intelligence system of human-like intelligent mainly lies in how to effectively imitate the approach with which human use imagery to cognize. From psychological point of view, the process of cognitive is the transformation process of objects and attributes. Based on this thought, imagery of conceptual network is constructed based on imagery concept and attribute to store the overall planning information of imagery object conceptual. According to imagery conceptual network we build, calculus method imagery of conceptual network based on revision calculus is proposed to guide filling and revision of the nodes of imagery of conceptual network. According to the instance calculus given in the paper can be seen, the calculus method imagery conceptual network can fill and revise the nodes of imagery conceptual network, which can effectively imitate storing imagery and induction imagery of the human brain.

Abstract: Efficient encoding of visual perception has a profound influence on machine vision and research and development of human-like intelligent systems. A coding method of visual perceptual reconstruction which was according to Gestalt perceptual organization principle was proposed. Visual perception connected discontinuous curves which can connect with reasonable smoothing method through curve fitting first before shape recognition and shape coding. Coding method filled in points of discontinuous place to get whole contour. Energy function was designed. Coding method can give the most reasonable solution when there are many methods of perception reconstruction. This paper described the implement method of coding method, and based on psychology used typical graphics which is used to explain the adjacent ratio to make experiment. The experiment results demonstrated that the coding method of visual contour reconstruction proposed accorded with Gestalt perception organization principle.

Abstract: Confirming the position in space of the camera axis and it perpendicular to the target surface has important significance for simplifying the camera calibration in interior parameters. This article conducted a study on this issue. First, it establishes the relationship of the world coordinate system, the target coordinate system, the camera coordinate system and the image coordinate system. The necessary camera calibration in external parameters and internal parameters are confirmed. Second, the first target coordinate system is adjusted to the same position with the world coordinate system. The adjustment principle of the camera axis perpendicular to the target column end surface is proposed. Namely, whether the camera axis perpendicular to the first target column end surface or not is confirmed by the CCD image edge whether it’s round or not. As the world coordinate system is based on coordinate measuring machines, so the world coordinate of the first target column end surface circle is obtained. Then the second target is placed in between the camera and the first target, and the second target coordinate system is adjusted to the same position with the world coordinate system, and it is adjusted to the same coordinate (except y direction) with the first target column end surface circle’s coordinates. So the images of the first target and the second target column end surface circle in the CCD surely are circles. Third, the method to seek for the camera axis is put forward, namely if the circles of the first target and the second target column end surface circle in the CCD are not concentric, then it shows the optical axis and the joined axis of the first and second target face round are not concentric. By adjusting the CCD camera's position, until the circles of the first target and the second target column end surface circle in the CCD are concentric, such the camera axis can be represented by the joined axis of the first and second target end round. Simultaneously the target circle's center reflects the deviation on CCD coordinate between the optical center and the CCD center. Finally, the camera internal parameters are expressed by the related parameters.

Abstract: Computer vision is a rapid, consistent and objective inspection technique, which has expanded into many diverse industries. Its speed and accuracy provide one alternative for an automated, non-destructive and cost-effective technique to accomplish ever-increasing production and quality requirements. This method of inspection has found applications in the agricultural industry, including the inspection and grading of fruits. This paper provides an introduction to main defection and grading approaches of fruit external defects, including image processing and pattern recognition methods based on fruit two-dimensional (2D) and three-dimensional (3D) information, and hyperspectral and multispectral imaging. In addition, their advantages and disadvantages are also discussed.

Abstract: A novel three dimensional reconstruction framework from wide baseline images was proposed based on point and line features. After detecting and matching features, the relations between discrete images are computed and refined according to multi-view geometric constraints, and both structure of the scene and motion of cameras are retrieved, where we employ a procedure of Euclidean reconstruction based on approximate camera internal parameters and buddle adjustments. Based on retrieved motion and correspondence of line features, a 3D line reconstruction scheme was put forward to assist us in gaining regular structure and topology of the scene. In virtue of some manual interactions, mesh models of the scene came into being, and a rectification method for perspective images was used to acquiring texture patches. Finally, an interactive modeling prototype system from multiple images is designed and implemented. Real scenes and augmented reality applications demonstrate the feasibility, correctness and accuracy of our framework.

Abstract: The internal radial tire defects will affect the performance and the service life of the automobile. But the traditional detection method is to observe and detect in eyes using X-ray map which result that a lot of quality problems can not properly been judged out. This paper describes the X-ray image recognition using a fuzzy algorithm. The improved algorithm simplifies the traditional fuzzy algorithm complexity without the transform and inverse transform. The experimental results show that the improved algorithm makes the computational efficiency and extract more refined and clear edge for the radial tire which can correctly identify the failure and provide the necessary data processing.

Abstract: Using TBD to reduce the minimum detectable SNR can be used to detect weak targets of pulsed laser radar sensor. Firstly, the pretreatment measures are given to enhance the single frame SNR. Especially, with the width characteristic of laser target under high sampling rate, the down-sampling method is introduced, which can not only eliminate a large number of single-point noises, but also can solve the acquisition problem of relatively stationary or low-speed targets. Then the TBD algorithm based on HT is proposed. Finally, the experiments indicate that the algorithm has good detection performance and can detect single and multiple targets.

Abstract: This paper gives an efficient VMIMO scheduling scheme for wireless sensor networks(WSNs) with a node selection algorithm. Scheduling communication scheme with feedback channel is proposed in this paper. Using the feedback channel, channel state information (CSI) could be return to transmitting cluster. Then the “better” node could be selected to participate in cooperation. Simulation results show the efficient of the proposed Scheme in energy saving.

Abstract: In background of the developing trend of the high-resolution remote sensing camera, the following paper presents some methods of reducing noise in signal processing circuit. Based on the theory of Signal Integrity, this study from several aspects, such as power filter, signal filter and etc, simulate and validate on a software named Cadence, in the end, verify on real circuit, and achieve very good results.