Papers by Keyword: Mahalanobis Distance

Paper TitlePage

Authors: Liang Zhang, Pei Yi Shen, Juan Song, Luo Bin Dong, Yan Zheng Zhang, Xiao Xi Zhang, Jie Qiong Zhang
Abstract: This paper proposes a new approach to the multi-robot map fusion algorithm that enables a team of robots to build a joint map without initial knowledge of their relative pose. First, the relative distance and bearing measurements between two robots are fused together by the covariance intersection method after they detect each other. Second, the transformation equations among multi robots coordinates are derived based on their relative distance and bearing measurements. Third, all the multi robots local maps are merged into one global map by unscented transform based on the transformation equations. Fourth, the possible duplicate features are filtered out by the robots maximal detection area and the features coordinate range, then the Mahalanobis distance is computed to decide the duplicate features correspondence through unscented transform, and the Kalman Filter is used while fusing the duplicate features information. As a means of validation for the proposed method, experimental results obtained from the two robots are presented.
Authors: Zhen Chen, Ji Hong Shen
Abstract: In this paper, a mahalanobis distance based flame fringe detection algorithm through digital image processing was proposed according to the insufficient accuracy and excessive interference of the traditional flame fringe detection algorithm. The similarity between the pixels in GRB image and the sample flame pixels was first calculated through Euclidean distance and mahalanobis distance for classifying the pixels in the image and finishing flame segmentation, and then the image was processed through binarization, and finally flame fringe was extracted through gradient method and image morphology. Also, a simulation analysis was made, and the results showed that the fringe extracted with this algorithm was single-pixel, smooth and continuous without cross, had less interference, and possessed high accuracy and reliability. Thus, this method can meet the flame detection in the complex images such as fire disaster image.
Authors: Yun Chi Yeh, Tsung Fu Chien, Cheng Yuan Chang, Tsui Shiun Chu
Abstract: This study proposes a Mahalanobis Distance Measurement (MDM) method to analyze current waveform for determining the motor’s quality types. The MDM method consists of three major stages: (i) the preprocessing stage which is for enlarging motor current waveforms’ amplitude and eliminating noises, and includes signal amplitude amplifier, filter circuit (eliminating noises), and analog-to-digital converter (ADC) parts, (ii) the qualitative features stage which is for qualitative feature selection on motor current waveforms, and (iii) the classification stage which is for determining motor quality types using the MDM method. It can recognize defective motors and their defective types in less than 0.5 second. In the experiment, the total classification accuracy (TCA) was approximately 99.03% in average. The proposed method has the advantages of good detection results, no complex mathematic computations, hi-speed, and hi-reliability.
Authors: Sheng Qiang Feng, Guo Ping Ling, Bing Ma, Sheng Sun Hu
Abstract: Online fault non-destructive detection is crucial to the fabrication of high quality products of welds. In this study, a novel method in the field of weld whose name is Mahalanobis Distance Measurement (MDM) is illustrated and employed to determine whether welding faults have occurred or not. This method is based on an arc method which is employed to get the information of welding voltage and welding current. Through analyzing the calculation results of Mahalanobis Distance of the welding voltage and current signals, the important information of welding process has been obtained. At the meanwhile, whether a fault has occurred or not is decided. In order to verify the effect of the performance of the different width of the gap on the results of Mahalanobis Distance, two widths of the gap generated artificially were chosen: one is 2 mm and the other is 3 mm. The results of the experiment have demonstrated that using the thinking of Mahalanobis Distance to measure the weld faults is feasible.
Authors: Kai Li, Yu Fei Zhou, Xiao Juan Li
Abstract: Semi-supervised clustering is a method which can improve clustering performance by introducing partial supervised information. This paper mainly studies the semi-supervised fuzzy clustering which introduces Mahalanobis distance and Gaussian Kernel. And we obtain a new semi-supervised fuzzy clustering objective function. By solving the optimization problem, we propose a semi-supervised fuzzy clustering algorithm F-SCAPC which includes F(M)-SCAPC and F(K)-SCAPC. And we do experimental research for proposed algorithm F-SCAPC using the selected standard data set and the artificial data set. Besides, we compare performance of presented algorithm F-SCAPC with AFFC, KFCM-F and SCAPC algorithms. From the results, we can see that F-SCAPC is effective in the convergence speed and the accuracy.
Authors: Qian Qian Wu, Ji Hye Lee, Jong Pyo Lee, Min Ho Park, Young Su Kim, Ill Soo Kim
Abstract: Gas Metal Arc (GMA) welding is considered as a multi-parameter process that it’s hard to find optimal parameters for good welding. To overcome the problem, an artificial neural network based on the backpropagation algorithm was built to realize the relationships between process parameters and welding quality as output parameter. In this study, Mahalanobis Distance (MD) was employed to evaluate the availability of a given welding parameters which was proved to performance well in multivariate statistics. Input parameters such as welding current and arc voltage were chosen due to their significant influence on the welding quality. To improve the precision of given parameters’ evaluation, neural networks with different configurations were verified. The analyses on the measured and predicted MD by the proposed neural network were conducted. The proposed neural network based on the error backpropogation algorithm was proved to have high reliability to evaluate process parameters, which further makes it available in on-line monitoring system.
Authors: Reenal Ritesh Chand, Ill Soo Kim, Ji Hye Lee, Jong Pyo Lee, Ji Yeon Shim, Young Su Kim
Abstract: In robotic GMA (Gas Metal Arc) welding process, heat and mass inputs are coupled and transferred by the weld arc and molten base material to the weld pool. The amount and distribution of the input energy are basically controlled by the obvious and careful choices of welding process parameters in order to accomplish the optimal bead geometry and the desired mechanical properties of the quality weldment. To make effective use of automated and robotic GMA welding, it is imperative to predict online faults for bead geometry and welding quality with respect to welding parameters, applicable to all welding positions and covering a wide range of material thickness. To successfully accomplish this objective, two sets of experiment were performed with different welding parameters; the welded samples from SM 490A steel flats adopting the bead-on-plate technique were employed in the experiment. The experimental results of current and voltage waveforms were used to predict the magnitude of bead geometry and welding quality, and to establish the relationships between weld process parameters and online welding faults. MD (Mahalanobis Distance) technique is employed for investigating and modeling of GMA welding process and significance test techniques were applied for the interpretation of the experimental data. Statistical models developed from experimental results which can be used to control the welding process parameters in order to achieve the desired bead geometry based on weld quality criteria.
Authors: Zhao Ming Shi, Bo Ying Geng, Zhong Hong Wu, Yin Wen Dong
Abstract: Aiming at problems about repeat matching and wrong matching appeared when traditional SIFT algorithm was used in image matching, an image matching method based on SIFT feature was put forward. Firstly, SIFT features were extracted by traditional SIFT algorithm and candidate matching point pairs were obtained by the nearest neighbor rule. Secondly, lateral matching method was used to remove repeat matched dot-pairs. Finally, Mahalanobis distance as a similarity measurement was used to remove wrong matched dot-pairs. Experiment shows this method can achieve image matching effectively with high accuracy.
Authors: Yin Wen Dong, Luan Wan, Zhao Ming Shi, Ming Lei Zhu
Abstract: Aiming at there are long matching time and many wrong matching in the traditional SIFT algorithm, An image registration algorithm based on improved SIFT feature is put forward. First of all, through setting the number of extreme points in the feature point detection, feature points is found according to the DOG space structure from coarse to fine, and the improved SIFT feature descriptor generation algorithm is used. The preliminary matched point pairs are obtained by the nearest neighbor matching criterion, and the bilateral matching method is used for screening the preliminary matched point. Then, the second matching will be done by the similar measurement method based on mahalanobis distance, and RANSAC algorithm is used to calculate the affine transform model. Finally, the transformed image is resampled and interpolated through the bilinear interpolation method. Experimental results show that the algorithm can realize image registration effectively. Image registration technique is an important research content in computer vision and image processing in the, which are widely used in vehicle matching navigation and positioning, cruise missile terminal guidance, target tracking and recognition, image mosaic[1-6]. SIFT algorithm[3-5] can achieve image registration when there are translation, rotation, affine transformation between two images, even for images took by arbitrary angles. And SIFT feature is the milestone of local feature study. But there are long matching time and many wrong matching in the traditional SIFT algorithm, it is difficult to meet the requirement of fast image registration. This paper puts forward an image registration algorithm based on improved SIFT feature, which is robust for image rotation, affine and scale change, and is better than traditional SIFT algorithm.
Authors: Mohd Yazid Abu, Khairur Rijal Jamaludin
Abstract: While the concept of remanufacturing, especially on automotive parts is gaining in popularity, in practice the remanufacturing industry in Malaysia is still in its nascent stage, with approximately 32 fields in various industries claiming to be involved in the process. The Mahalanobis-Taguchi System (MTS) is a diagnostic method employing Mahalanobis Distance (MD) for recognizing different patterns in multivariate data. The aim of this work is to apply T method-3, which is one of the sub-methods under the MTS relating to the main journal diameter of the crankshaft. The method distinguishes between two distinct ranges of acceptable remanufacturing and non-remanufacturing processes. Furthermore, the method also categorizes various patterns of crankshaft based on their MD in unit space. The case study is performed in an automotive industry in Malaysia under a contract remanufacturing environment. The outcome of this work is expected to be the enhancement of the robustness of the remanufacturing system on pattern recognition to the company under study. As a result, the company is expected to save more time and energy in coming with faster decision-making. In addition, the study would provide greater inspiration, especially among researchers in aggressively applying MTS applications to a wider variety of industry sectors especially in the remanufacturing area.
Showing 1 to 10 of 25 Paper Titles