Advanced Materials Research Vols. 139-141

Abstract: In accurate optical inspection of IC chip, it is necessary to recognize and calibrate the IC micro-topography completely and accurately. Positioning method of characteristic points in IC chip’s micro-topography based on cross correlation matching and image projection transformation is investigated to realize high precision detection of surface topography. On the basis of extracting wavelet high-frequency components of IC topography conjugate images which is shot by high- definition CCD cameras from different space positions, the coordinate information of some selected topography’s characteristic points is shown. Cross correlation matching is applied to definite the characteristic point’s relative positions in the projected images, then relationship of image projection transformation is established for calculating and determining the characteristic point’s absolute coordinates, which facilitates following topography detection of IC chip. Experiment analysis and data comparison indicate that accurate coordinate positioning results of characteristic points can be obtained by this method, the research basis for following IC topography model structuring and micro- flaw characteristic detecting can also be provided.

Abstract: Crack is the common drawback in deposition manufacturing process with high energy beam, which usually leads to the manufacturing process being given up halfway. So it is important and urgent to investigate the generating principle of crack and establish relative means to prevent crack appearance. In this paper, the principle of crack generating in the process of hybrid plasma and laser manufacturing was studied experimentally and systematically. The influences of power system and shielding atmosphere on crack were researched. The results were summarized as follows: (1) The crack generating in the course of manufacturing is mainly heat crack. (2) Excessive energy density and temperature will result in crack. (3) The best method preventing part being manufactured from oxidation is to put the whole part into the inert gas.

Abstract: In order to improve the classification rate of gait recognition, a new gait recognition algorithm is proposed. Firstly, the gait images are preprocessed, and the outlines of gait images are extracted and normalized. Secondly, wavelet moments of the outlines are calculated to describe the static feature of the gait images. Thirdly, the leg double triangle model is built. The first triangle consists of the mid-point of the two hips, left knee point and right knee point, and the other one consists of the mid-point of the two hips, left ankle point and right ankle point. Then the parameters of two triangles are extracted to describe the dynamic features of the gait images. Finally, the above two features are fused and used for the classification. The experimental results show that proposed algorithm provides higher correct classification rate than the algorithms using single feature, and meets the requirements of the real-time.

Abstract: A new method of actualizing grating non-diffracting structured light (GNSL) is proposed. The GNSL holds characters of long focal-depth, narrow strip-width, steady light field and constant light intensity in certain rang etc. Based on precise triangular-section prism, an optical system is designed and actualized experimentally. With theory analysis and calculation to the optical system, the distribution of the light intensity is simulated in non-diffracting range. Experiment results show that simulation is in good agreement with the theory calculation. According to system parameters, GNSL with 18um strip-width and 57cm non-diffracting range is produced in experimental system. Compared with traditional projected light source, GNSL has advantages of high resolution and high precision in the application of three dimensions (3D) shape measurement. By using GNSL, matching ambiguity could be reduced.

Abstract: A research on measurement method of space position in three-dimensional field was demonstrated in this paper. According to the basic methods and principles of three-dimensional spatial position measurement, optical path of splitting design was used to divide an object point, through the red and green filters, into two beams of red and green, and then, through flat mirror, rectangular prism, cube prism and other optical design, it imaged on the same CCD camera film. Standard mesh was used to determine the relationship between two different spatial coordinates and the spatial pixel coordinates in the measurement space, and then the calibration relationship between object point coordinates and pixel coordinates were determined. This calibration relationship clear the error caused by the method of measurement and the measurement system. Through the determination algorithm of spatial location in three-dimensional field, the two corresponding lines of space was obtained, and its point of intersection is the spatial coordinate.

Abstract: To segment the pebrine from the background effectively, this paper focuses on the research of image segmentation technology on the basis of HSI color model. Owing to the complex background and poor contrast of the pebrine image, the method of contrast enhancement is adopted to improve the image quality; the technology of target extraction based on HSI model is used to realize the direct extraction of the colorized object image from the background, which utilizes the color characteristic of pebrine image; Meanwhile, the research on the threshold processing on the basis of shape feature of pebrine image can be done to improve the efficiency of algorithm, which minimizes the influence of non-target image, and the method of morphology can be applied to eliminate the hole or dot noise, and to separate the adhesive pebrines, which increases the accuracy of pebrine recognition. The result shows the method reasonable and effective.

Abstract: With the safety awareness strengthened, identification authentication technology has been increasingly concerned. Face recognition is attractive in pattern recognition and artificial intelligence field, and face feature extraction is a very important part in face recognition. This paper first introduced preprocessing of face images, PCA and ICA algorithm. Considering PCA and ICA their respective strengths and weaknesses, then a novel face feature extraction method based on PCA and ICA is proposed. The NN classifier is select to face classification and recognition on the ORL face database. From the actual requirements, the paper analyses hardware platforms based on DM642, and finally use tool CCS software to optimize program and implementation base on DM642 to meet the real-time requirements. Experiments indicated that the modified method is superior to PCA and ICA algorithm.

Abstract: The human pulse-condition diagnosis is an important part of the traditional Chinese medicine (TCM) which is difficult to recognize accurately by doctor’s subjective experience. Objective identification of pulse-conditions has important meanings for modernization of TCM. In this paper human pulse-condition system transfer function model and model parameter estimation were introduced, which are used to construct four kinds of typical pulse-conditions simulation signals. There are normal pulse, taut pulse, slippery pulse and thready pulse. And then, discrete wavelet transform for extracting the multi-scale energy characteristics and wavelet packet decomposition for extracting the multi-band energy characteristics are proposed so as to recognize the pulse-conditions simulation signals. The results show that the recognition effect of discrete wavelet transform method is better. Moreover, the data features of characteristic parameters demonstrate the reality of simulation signals.

Abstract: Flatness is one fundamental element of geometric forms, and the flatness evaluation is particularly important for ensuring the quality of industrial products. This paper presents a new flatness evaluation in the view of the minimum zone evaluation - rotation method based on genetic algorithm. This method determines the minimum zone through rotating measurement points in three dimensions coordinate. The points are firstly rotated about coordinate axes. Then they are projected in one axis, and the smallest projection length is the flatness value. The rotation angles are optimized by genetic algorithm to improve search efficiency. An exponential fitness function and the rotation angles range is designed on the basis of flatness characteristics. An adaptive mode of crossover and mutation probability is used to avoid local optimum. The results show this method can search the minimum zone and converge rapidly.

Abstract: A detection device was designed to detect the poultry meat edible quality. The detection device is the electromechanical integration equipment which is integrated with x-ray technology, visible light technology and induced fluorescence technology. The device not only can overcome the disadvantages that the conventional measure methods have destructiveness to the samples, and consume the long time of the detection, etc, but also can detect the multiple quality indexes simultaneously. The detection device is composed of the frames, the transmission control system, the X-ray detection system, the visible light detection system and the induced fluorescence detection system, etc. The device can greatly increase the judging ability for comprehensive quality of poultry meat and realize the fast and non-destructive testing to poultry meat edible quality.