Advanced Materials Research Vols. 889-890

Abstract: For the floor type boring-milling machining center, when the ram and the spindle protrude from the headstock, a cantilever-like structure is formed which will lead to great stiffness error and will affect the machining accuracy of the workpiece without compensation. It has been a difficult problem to fulfill the compensation of this stiffness error. Regarding this issue, a new method is discussed in this paper to realize this kind of compensation. By combining linear superposition principle, optimization theory with the FEA analysis, a formula is presented to calculate the oil pressure of the compensating system, which describes the relationship between the oil pressure of the compensating hydraulic cylinder and the stroke of the ram. Therefore, the oil pressure of the compensating system can be accurately calculated.

Abstract: The technology development of the large gantry machine has been more and more popular, which is on the basis of the development of modern processing method and generalization of complex products. In order to reduce non-cutting time and improve machining efficiency, a large capacity of chain tool library was introduced in this paper, the structure of the whole chain tool library and tool changer were designed firstly, and then the simulation of automatic tool changer was conducted in the UG to verify the correctness of the movement. Finally, the design parameters were optimized according to the simulation results. And we can obtain the most appropriate motion rule by adjusting design parameters in the process of digital machining simulation.

Abstract: In order to solve the problem of safe automation ignition process for vanadium iron smelting furnace, characteristics of ignition process, matching problem of device to process and its security design direction were analyzed. The requirements of process safety improvement were proposed. Key designs includes ignition furnace trolley positioning mechanism, swinging arm mechanism, the fire pole stretching mechanism, safety parclose, security monitoring devices, automated process of electronic ignition and automatic withdrawal of ignition guns. Application shows 13 safety advantages. In the research, process analysis is foundation, it includes steps refinement of manufacturing operations process, the order, the status, and problems that may occur in each step, this method for automation safe ignition process building could be a reference for similar research and development.

Abstract: Image fusion technology is widely used on digital image processing to improve the quality of target image that are included in a series partial focusing images. In this paper, we proposing a region weighted wavelet fusion method, which is also considering the neighborhood region of interest point. The comparison of new method and common wavelet fusion method demonstrate that new method will improve the fusion image quality with less wavelet decomposition level.

Abstract: This article will use visual inspection techniques into the two-dimensional geometry measurement of ceramic products .It uses MATLAB 7.0 as image processing software, and the designing uses the gradient information into detection circular edge by Hough transformation method. The method is helpful to improve the production efficiency and the recognition rate. We can prove that this method has the feasibility by experiments, and achieve a satisfactory result .

Abstract: To Solve how to Calibrate Rapidly and Accurately in Camera, it Proposed the Method of Hyperchaos Least Square and Implemented by Labview through Analyzing the Camera Model and the Calibration Principle. it is that Measuring Image Coordinate and World Coordinate, then Calculating the Important Camera Internal and External Parameters which are R Rotation Matrix and T Translation Matrix Solutions Based on Hyperchaos Least Square Method.

Abstract: To improve the pertinence and accuracy of grating image barcode detection, a method based on Hough transform and morphological operation about grating image full-barcode correction and extraction is proposed. The basic principles of Canny edge detection, Hough transform and morphological operation are introduced. Non full-barcode can be eliminated after Canny detection and morphological operation. The level correction can be finished by using Hough transform to detect the full-barcode edge lines and get the correction angle. The full-barcode extraction can be achieved after the dilation of correction grating image and the reduction of free defects. The experimental results show that not only the non full-barcodes and free defects can be better filtered, but also we can accurately accomplish the correction and extraction of the full-barcodes.

Abstract: The objective of this work is to discuss the Evaluation method of material surface corrosion based on image recognition of computer technology. The color information and texture of the material surface corrosion is obtained and digitalized by the CCD camera, and are processed by the computer; the indexes of color information is extracted based on the color model and texture features are extracted from the corrosion images based gray occurrence matrix. Those features are compared to the features of standard image of each kind of material surface corrosion grade in image database. Finally, the material surface corrosion grade is defined. The research implies that the computer recognition technology could be used to evaluate the corrosion grade of material surface corrosion.

Abstract: One method named image synthesis was proposed for gray image composition to meet the requirement that algorithm research of machine vision needed a lot of images. Firstly, two modules, i.e., modeling module and camera module, of some engineering software were compared and images will be obtained by using three-dimensional modeling software cooperating with VRay plug-in. Then, parameter setting and adjustment for VRay plug-in was studied. Lastly, measurement experiments were carried out by using synthetic images. The experimental results demonstrate that 1μm measuring accuracy is possible under the condition that the distance between calibration part and camera was equal to that between parts and camera. The synthetic images can be used for algorithm research of machine vision.

Abstract: Human action often requires a large volume and computation-consuming representation for an accurate recognition with good diversity as the large complexity and variability of actions and scenarios. In this paper, an efficiency combined action representation approach is proposed to deal with the dilemma between accuracy and diversity. Two action features are extracted for combination from a Kinect sensor: silhouette and 3D message. An improved Histograms of Gradient named Interest-HOG is proposed for silhouette representation while the feature angles between skeleton points are calculated as the 3D representation. Kernel Principle Componet Analysis (KPCA) is also applied bidirectionally in our work to process the Interest-HOG descriptor for getting a concise and normative vector whose volume is same as the 3D one aimed at a successful combining. A depth dataset named DS&SP including 10 kinds of actions performed by 12 persons in 4 scenarios is built as the benchmark for our approach based on which Support Vector Machine (SVM) is employed for training and testing. Experimental results show that our approach has good performance in accuracy, efficiency and robustness of self-occlusion.