• Journal of Computing Science and Engineering
• /
• v.11 no.1
• /
• pp.1-8
• /
• 2017

Naive Bayes nearest neighbor (NBNN) is a simple image classifier based on identifying nearest neighbors. NBNN uses original image descriptors (e.g., SIFTs) without vector quantization for preserving the discriminative power of descriptors and has a powerful generalization characteristic. However, it has a distinct disadvantage. Its memory requirement can be prohibitively high while processing a large amount of data. To deal with this problem, we apply a spherical hashing binary code embedding technique, to compactly encode data without significantly losing classification accuracy. We also propose using an inverted index to identify nearest neighbors among binarized image descriptors. To demonstrate the benefits of our method, we apply our method to two existing NBNN techniques with an image dataset. By using 64 bit length, we are able to reduce memory 16 times with higher runtime performance and no significant loss of classification accuracy. This result is achieved by our compact encoding scheme for image descriptors without losing much information from original image descriptors.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.11a
• /
• pp.251-256
• /
• 2004

The image classification is one of the most important studies in the remote sensing. In general, the MLC(Maximum Likelihood Classification) classification that in consideration of distribution of training information is the most effective way but it produces a bad result when we apply it to actual hyperspectral image with the same classification technique. The purpose of this research is to reveal that which one is the most effective and suitable way of the classification algorithms iii the hyperspectral image classification. To confirm this matter, we apply the MLC classification algorithm which has distribution information and SAM(Spectral Angle Mapper), SFF(Spectral Feature Fitting) algorithm which use average information of the training class to both multispectral image and hyperspectral image. I conclude this result through quantitative and visual analysis using confusion matrix could confirm that SAM and SFF algorithm using of spectral pattern in vector domain is more effective way in the hyperspectral image classification than MLC which considered distribution.

• Journal of the Chungcheong Mathematical Society
• /
• v.25 no.2
• /
• pp.367-377
• /
• 2012

The goal of the image restoration is to find a good approximation of the original image for the degraded image, the blurring matrix, and the statistics of the noise vector given. Fast truncated Lagrange (FTL) method has been proposed by G. Landi as a image restoration method for large-scale ill-conditioned BTTB linear systems([3]). We implemented FTL method for the image restoration problem with reflective boundary condition which gives better reconstructions of the unknown, the true image.

• Journal of KIISE:Software and Applications
• /
• v.26 no.1
• /
• pp.150-150
• /
• 1999

In this paper, a content-based image retrieval scheme based on scale-space theory is proposed. The existing methods using scale-space theory consider all scales for image retrieval,thereby requiring a lot of computation. To overcome this problem, the proposed algorithm utilizes amodified histogram intersection method to select candidate images from database. The relative scalebetween a query image and a candidate image is calculated by the ratio of histograms. Feature pointsare extracted from the candidates using a corner detection algorithm. The feature vector for eachfeature point is composed of RGB color components and differential invariants. For computing thesimilarity between a query image and a candidate image, the euclidean distance measure is used. Theproposed image retrieval method has been applied to various images and the performance improvementover the existing methods has been verified.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.6
• /
• pp.895-906
• /
• 1993

In this paper, MTSVQ(Multirate Tree-Structured Vector Quantization) is introduced for subband image coding. Original images are decomposed into a number of subband components, and multiresolution codebook is designed by MTSVQ algorithm. Optimal bit allocation among the subband components becomes the problem selecting the particular pruned subtree of MTSVQ which has the desired rate and distortion.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.1
• /
• pp.420-427
• /
• 2017

This paper proposes a new method using Wide Line Detector based statistical shape Feature (WLDF) to identify whether or not a tongue is cracked; a cracked tongue is one of the most frequently used visible features for diagnosis in traditional Chinese Medicine (TCM). We first detected a wide line in the tongue image, and then extracted WLDF, such as the maximum length of each detected region, and the ratio between maximum length and the area of the detected region. We trained a binary support vector machine (SVM) based on the WLDF to build a classifier for cracked tongues. We conducted an experiment based on our proposed scheme, using 196 samples of cracked tongues and 245 samples of non-cracked tongues. The results of the experiment indicate that the recognition accuracy of the proposed method is greater than 95%. In addition, we provide an analysis of the results of this experiment with different parameters, demonstrating the feasibility and effectiveness of the proposed scheme.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.9 no.2
• /
• pp.144-150
• /
• 2000

Surface roughness is one of the most important parameters to estimate quality of products. As this reason so many studies were car-ried out through various attempts that were contact or non-contact using computer vision. Even through these efforts there were few good results in this research., however texture analysis making a important role to solve these problems in various fields including universe aviation living thing and fibers. In this study feature value of co-occurrence matrix was calculated by statistic method and roughness value of worked surface was classified, of it. Experiment was carried out using input vector of neural network with characteristic value of texture calculated from worked surface image. It's found that recognition rate of 74% was obtained when adapting texture features. In order to enhance recogni-tion rate combination type in characteristics value of texture was changed into input vector. As a result high recognition rate of 92.6% was obtained through these processes.

• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.40-47
• /
• 2003

In this paper, we propose a new method for reducing the computational overhead of fine-to-coarse multi-resolution motion estimation (MRME) at the finest resolution level by searching for the region to consider motion vectors of the coarsest resolution subband. At this time, if half-pel accuracy motion estimation (HPAME) is used in the baseband where influence a lot of effect to the reconstructed image, we can have the motion vector exactly But, this method causes to higher computational overhead. So we suggest the method to the computational overhead by using selective interpolation. Experimental results show that the proposed algorithm gives better results than the traditional algorithms from image quality.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.543-546
• /
• 2007

Flow visualization is one of visualization techniques and it means a visual expression of vector data using 2D or 3D graphics. It aims for human to easily find and understand a special feature of the vector data. The Image Based Flow Visualization (IBFV) is one of the fastest technique in the dense integration based flow visualization techniques. In this paper, IBFV is accelerated and implemented using commodity GPU. Especially, mesh advection is accelerated at the vertex program.

• Proceedings of the KIEE Conference
• /
• 1987.07b
• /
• pp.1438-1440
• /
• 1987

In this paper, we discussed the quantization technique in the DCT domain employing a vector quantizer (VQ), and described the relations between the DCT and the VQ. And, we introduced a zonal coding technique for the OCT coefficients based on the classified VQ technique proposed in [2]. We shall show that this technique reduced the coding complexity about 30% while maintaining the same image quality as shown in [2]. A result of simulation with a natural image is also presented.