Shape Modification from Endoscope Images by Regression Analysis

Yuji Iwahori,Seiya Tsuda,Aili Wang,Robert J. Woodham,M. K. Bhuyan,Kunio Kasugai 보안공학연구지원센터 2015 International Journal of Multimedia and Ubiquitous Vol.10 No.9

The VBW (Vogel-Breuß-Weickert) model is proposed as a method to recover 3-D shape under point light source illumination and perspective projection. However, the VBW model recovers relative, not absolute, shape. Here, shape modification is introduced to recover the exact shape. Modification is applied to the output of the VBW model. First, a local brightest point is used to estimate the reflectance parameter from two images obtained with movement of the endoscope camera in depth. A Lambertian sphere image is generated using the estimated reflectance parameter and VBW model is applied for a sphere. Regression analysis is introduced to improve the surface gradients, where linear coefficients can be obtained using true values of gradient parameters with a generated sphere. Depth can then be recovered using the modified gradient parameters. Performance of the proposed approach is confirmed via computer simulation and real experiment.

SVM Based Defect Classification of Electronic Board Using Bag of Keypoints

Hidenobu Inoue,Yuji Iwahori,Boonserm Kijsirikul,M. K. Bhuyan 대한전자공학회 2015 ITC-CSCC :International Technical Conference on Ci Vol.2015 No.6

This paper proposes a new approach for the defect classification of electronic board using Bag of Keypoints and SVM. The main purpose of this paper is not to use the reference image which can be used to extract the difference region of defect. The approach represents histogram features of Bag of keypoints based on extracting features from data set images. Feature vectors are used for SVM learning and classification. The effectiveness of the approach is evaluated with accuracy of defect classification for images with actual defects in comparison with the previously proposed approaches.

A Novel Image Superresolution Reconstruction Algorithm Based on Sparse Representation

Aili Wang,Xinyuan Wang,Yuji Iwahori,Yuan Feng,Na Jiang 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.6

Superresolution image reconstruction technique uses single or a series of low-resolution images to reconstruct a high resolution image without changing the hardware devices, while improving image quality and the spatial resolution of the image. High resolution means the image with a higher pixel density, can provide more details. In this paper, a novel image superresolution algorithm based on sparse representation is studied. During over-complete dictionary of the training phase, the proposed method improves two aspects including feature extraction and dimension reduction. In the feature extraction process, combining the second derivative with the gradient direction, we construct a new descent direction to improve gradient method. The convergence speed of the new algorithm is faster than the gradient method and can get better results. Then improved two-dimensional Principal Component Analysis (2DPCA) algorithm is used to reduce the dimension, it could eliminate the correlation of the image lines and column. Experiment results show that this method of image reconstruction is better and faster for high resolution image reconstruction.

Medical Image Fusion in NSCT Domain Combining with Compressive Sensing

AiliWang,Jiaying Zhao,Yuji Iwahori,Shiyu Dai 보안공학연구지원센터 2015 International Journal of Multimedia and Ubiquitous Vol.10 No.5

In recent years, with the development of compressive sensing (CS) theory, it has been widely applied to each field including image fusion, and obtained better fusion effect. And CS can reduce dimensions and the amount of data characteristics as well as the large amount and high computation complex. Therefore, this paper proposes a novel medical image fusion method based on compressive sensing theory in non-subsampled contourlet transform (NSCT) domain. First, NSCT transform is applied to the source images, and the coefficients in low frequency subband are fused by mean rules. For high frequency subband, CS is applied and the coefficients are fused by neighborhood-energy-MAX (NE-MAX) rule, then inverse CS is used to get fused coefficients. Finally, inverse NSCT is applied to get the reconstructed image. The experimental results show that the fusion algorithm proposed in this paper is superior to fusion method based on WT-MAX and CS-MAX、CS-MEAN.

A Novel Feature Detection Algorithm Based on Improved 2DPCA- SIFT

AiliWang,Yangyang Zhao,Jiaying Zhao,Yuji Iwahori,Xinyuan Wang 보안공학연구지원센터 2015 International Journal of u- and e- Service, Scienc Vol.8 No.6

Stable local feature and representation is a fundamental component of many image registration, 3D reconstruction and object recognition algorithms. SIFT is a good descriptor that encodes the salient aspects of the image gradient in the feature point’s neighborhood. This paper improved SIFT- based local image descriptor and proposed a SIFT feature matching algorithm based on improved 2DPCA which can eliminate both rows and columns of relevance. Experimental results show that improved 2DPCA-SIFT algorithm is relatively stable, accurate and fast.

Medical Image Fusion based on Pulse Coupled Neural Network Combining with Compressive Sensing

AiliWang,Jiaying Zhao,Shiyu Dai,Yuji Iwahori,Yangyang Zhao 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.5

Image fusion is an important branch of information fusion, widely used in various fields, especially in medical field. So increasing the quality and efficiency of medical image fusion has great significance. Combining the advantages of pulse coupled neural networks with Compressive Sensing; this paper puts forward a novel image fusion method in NSCT transform domain. First, NSCT transform is applied to the source images, and the coefficients in low frequency coefficient are fused by mean rules. For high frequency coefficient, CS is applied and PCNN. Finally, inverse NSCT is applied to get the reconstructed image. The experimental results show that the fusion algorithm proposed in this paper in the performance and integration efficiency has better fusion results.

Pedestrian Detection Algorithm Combining HOG and SLBP

Aili Wang,Mingxiao Wang,Jitao Zhang,Yuji Iwahori,Bo Wang 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.10

In order to solve the problem of pedestrian detection performance, the described operator was improved. In this paper, semantic local binary pattern (SLBP) and histogram of oriented gradient (HOG) are combined as new feature operator. This feature method would enrich the information and enhance the detection performance. And then histogram intersection kernel support vector machine (HIKSVM) classifier is trained by the augment feature. Because the time cost is too large by the conventional SVM. HIKSVM could make up this drawback, and significantly reduce the training time. The experiments on the INRIA pedestrian dataset show that the method obtained significant improvement in accuracy comparing to HOG descriptors.

3D Reconstruction of Remote Sensing Image Using Region Growing Combining with CMVS-PMVS

Aili Wang,Na An,Yangyang Zhao,Yuji Iwahori,Rui Kang 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.8

The characteristics of remote sensing image is not obvious and can not reflect the be reconstructed the detail characters of the objects. For the sparse points in multi images fusion results, texture selection and it is relatively difficult problems, for accurate reconstruction of remote sensing image details, this paper presents CMVS-PMVS(cluster multi view stereo- patch based multi view stereo) combining with region growing for computing dense matching points. First, select some seed points by region growing algorithm, and find the matching relationship between the seed points, followed by the matching relation from the seed point to spread until to the entire image. Then the image set are clustered by CMVS in order to reduce the amount of data in the process of reconstruction, and the operation rate and reconstruction accuracy can be improved. PMVS reconstruction method is used to complete the reconstruction task by matching, expanding and filtering three steps. The experimental results showed that the point cloud is dense enough which are reconstructed by the 3d reconstruction algorithm based on regional growth combining CMVS-PMVS and well expressed the practical model of object reconstruction, the reconstruction of objects in remote sensing images has very strong practicability.

Blocking Variable Step Size Forward-Backward Pursuit Algorithm for Image Reconstruction

AiliWang,Mingji Yang,Xue Gao,Yuji Iwahori 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.2

Compressed sensing is a new signal sampling theory that fully makes use of signal’s sparsity or compressibility. The theory shows that, the acquisition of a small amount of the sparse or compressible signal value can be used for exact signal reconstruction. Based on the study and summarization of the existing reconstruction algorithms, this paper proposes a novel blocking variable step size forward-backward pursuit (BVSSFBP). This paper proposed variable step size forward-backward pursuit algorithm by introducing the concept of sparse phase and variable step size to deal with different situations. The algorithm also divides two-dimensional image into blocks, in order to reduce the scale of observation matrix during single processing, reduce the single processing speed and the overall running time. Experimental results show BVSSFBP algorithm can obtain better reconstructed image quality.