Earticle

지역 패턴을 정확하게 부호화 하는 방법은 텍스처 분류 연구에 매우 중요한 요소다. 하지만 기존 널리 연구된 LBP기반 방 법들은 잡음에 취약한 근본적인 문제점이 있다. 최근 표정인식 분야에서 에지반응 값과 방향 정보를 활용한 LDP방법이 제안 되었다. LDP방법은 LBP보다 잡음에 강하고 더 많은 정보를 코드에 수용할 수 있는 장점이 있지만 텍스처 분류에 적용하기에 는 치명적인 회전 변화에 민감한 단점이 있다. 본 논문에서는 LDP 방법에 회전 불변 특성을 결합하고 기존 LDP가 가지고 있 던 부호 정보를 수용하지 않은 단점과 밝기 값 차이가 적은 영역에서 의미 없는 코드가 생성되는 단점을 극복한 새로운 지역 패턴 부호화 방법인 Rotation Invariant Local Directional Pattern 방법을 제안한다. 본 논문에서 제안된 방법의 텍스처 분류 성능을 입증하기 위해 널리 사용되는 UIUC, CUReT 데이터 셋에서 텍스처 분류를 수행했다. 그 결과 제안된 RILDP방법이 기존 방법보다 우수한 성능을 보여주었다.

Accurate encoding of local patterns is a very important factor in texture classification. However, LBP based methods w idely studied have fundamental problems that are vulnerable to noise. Recently, LDP method using edge response and dire ction information was proposed in facial expression recognition. LDP is more robust to noise than LBP and can accommod ate more information in it's pattern code, but it has drawbacks that it is sensitive to rotation transforms that are critical to texture classification. In this paper, we propose a new local pattern coding method called Rotation Invariant Local Direc tional Pattern, which combines rotation-invariant transform to LDP. To prove the texture classification performance of the proposed method in this paper, texture classification was performed on the widely used UIUC and CUReT datasets. As a result, the proposed RILDP method showed better performance than the existing methods.

Objective: This study aimed to develop texture-modified soups and classify them based on the International Dysphagia Diet Standardization (IDDSI) and National Dysphagia Diet (NDD) criteria. Methods: Ten soups were selected, and a xanthan gum-based thickener was added to adjust their viscosity. Flow tests with a 10 ml syringe and viscosity measurements using a rheometer were carried out according to the IDDSI and NDD criteria. Results: The addition of 1 g of thickener to the selected commercial soups resulted in different viscosity levels depending on the soup type. Under the IDDSI framework, seven soups (R SMG, BMG, KHG, SLT, B SUG, BGG, DGT) were categorized as level 1, two soups (B MYG, ADG) as level 2, and one soup (R DJJ) as level 3. As per NDD guidelines, seven soups (R SMG, BMG, KHG, SLT, B SUG, BGG, DGT) were classified as Nectar-like and three soups (R DJJ, B MYG, ADG) as Honey-like. However, the addition of 2 g of thickener decreased the fluidity of the soups, with two soups (R SMG, SLT) being classified as level 2 and eight soups (R BMG, DJJ, KHG, B SUG, MYG, ADG, BGG, DGT) as level 3 under the IDDSI criteria. Meanwhile, all soups were classified as Honey-like under the NDD criteria. Conclusion: The viscosity was different due to the varying compositions of guk/tang/jjigae, the concentration of the thickener, and time. Swallowing standardsands and other guidelines applicable specifically to Korean-food for dysphagia patients need to be developed in a further study.

본 논문에서는 텍스쳐와 컬러 정보를 기반으로 비디오 감시를 위한 빠른 물체 분류 방법을 제안한다. 영상들로부터 SURF와 색 히스토그램의 국부적 패치들을 추출하여 그들의 장점을 이용한다. SURF는 명암 내용 정보를 제공하고 색 정보는 패치에 대한 특이성을 증강시킨다. SURF의 빠른 계산뿐만 아니라 객체의 색 정보를 활용한다. 국부적 특징을 이용하여 관심 영역 혹은 영상의 전역적 서술자를 생성하기 위해 Bag of Word 모델을 이용하고, 전역적 서술자를 분류하기 위해 Na$\ddot{i}$ve Bayes 모델을 이용한다. 또한 본 논문에서는 판별적인 기술자인 SIFT도 성능 분석한다. 네 종류의 객체에 대한 실험결과 95.75%의 인식률을 보였다.

In this paper, we propose a fast object classification method based on texture and color information for video surveillance. We take the advantage of local patches by extracting SURF and color histogram from images. SURF gives intensity content information and color information strengthens distinctiveness by providing links to patch content. We achieve the advantages of fast computation of SURF as well as color cues of objects. We use Bag of Word models to generate global descriptors of a region of interest (ROI) or an image using the local features, and Na$\ddot{i}$ve Bayes model for classifying the global descriptor. In this paper, we also investigate discriminative descriptor named Scale Invariant Feature Transform (SIFT). Our experiment result for 4 classes of the objects shows 95.75% of classification rate.

사람은 독서나 필기 중 중요 문구를 형광펜으로 칠하는 것에서 착안하여, 본 논문에서는 복잡한 배경 질감을 가진 영상에서의 불량유형을 효과적으로 분류하기 위해 불량 픽셀 영역을 하이라이팅 하여 신경망을 훈련하는 방법을 제안한다. 제안 방법의 가능성을 검증하기 위하여 불량유형 구분이 매우 어려운 타이어 밴드 직물의 불량유형 분류에 제안 방법을 적용한다. 또한, 타이어 밴드 직물 영상에 특화된 백라이트 하이라이팅 방법을 제안한다. 백라이트 하이라이트 영상은 GradCAM 기법과 간단한 영상처리를 이용하여 획득할 수 있다. 실험에서 우리는 제안하는 하이라이팅 기법이 분류 정확도뿐만 아니라 훈련속도 면에서 기존 방법보다 우수함을 보였다. 인식률 면에서는 제안 방법이 기존 방법 대비 최대 13.4%의 향상을 달성하였다. 타이어 밴드 직물 영상에 특화된 백라이트 하이라이팅 기법이 윤곽 하이라이팅 기법보다 정확도 측면에서 우수함을 보였다.

Motivated by people highlighting important phrases while reading or taking notes we propose a neural network training method by highlighting defective pixel areas to classify effectively defect types of images with complex background textures. To verify our proposed method we apply it to the problem of classifying the defect types of tire band fabric images that are too difficult to classify. In addition we propose a backlight highlighting technique which is tailored to the tire band fabric images. Backlight highlighting images can be generated by using both the GradCAM and simple image processing. In our experiment we demonstrated that the proposed highlighting method outperforms the traditional method in the view points of both classification accuracy and training speed. It achieved up to 13.4% accuracy improvement compared to the conventional method. We also showed that the backlight highlighting technique tailored for highlighting tire band fabric images is superior to a contour highlighting technique in terms of accuracy.

Textural patterns can often be used to recognize familiar objects in an image or retrieve images with similar texture from a database. Texture patterns can provide significant and abundance of texture and shape information. One of the recent significant and important texture features called Texton represents the various patterns of image which is useful in texture analysis. However sometimes the textured image obtained may not be of good quality and this may lead to improper detection of significant patterns. To enhance the quality or better illumination or contrast or sharpening effect, the present paper applied various local grey to grey level preprocessing steps on textured data. Grey to grey level preprocessing is required because the patterns on textons can be evaluated only on grey level textures. The present paper evaluated the occurrence behavior of various texton patterns based on the various grey to grey level pre processing methods, for an efficient rotationally invariant texture classification. The experimental results on various stone textures indicate the efficacy of the proposed method when compared to other methods.

A new feature set derived from the fractal geometry, called the random threshold vector (RTV) is proposed for texture analysis. The RTV is computed for different run length entropy dimensions. The run length entropy dimensions are calculated based on different thresholds. To test the rotationally invariant feature, the run length entropies are calculated in different directions. The experimental results show that the RTV contains great discriminatory information needed for a successful classification.

Study of different patterns on a local neighborhood of a texture plays an important role in characterization, and classification of the textures. The present paper proposes a method for measuring the occurrence factor of patterns on a randomly thresholded binary image. For this process eight simple patterns are chosen on a 3×3 neighborhood. The simple patterns are chosen in such a way that any complex pattern can be formed by grouping one or more of these simple patterns. The pattern occurrence factor of different binary images is also compared with the actual binary texture image. The experimental results on sixty four textures indicate good comparison of variation of occurrence in these patterns on different binary images of random threshold

Texture classification is important step in image processing and computer vision applications. The proposed method offers efficient way to classify the invariant texture using discrete shearlet transform and fuzzy logic. The texture features of an image are represented using shearlet energy features and shearlet co-occurrence features. These features are obtained from block based energy form of shearlet decomposed image using two levels of discrete shearlet transform with two directions and by varying the block size. Finally, the obtained parameters are used to classify the texture in an image using fuzzy logic classifier.

Identifying the specific machining processes used to produce specific workpiece surfaces is very useful in materials inspection. Machine vision can be used to semi- or fully automate this identification process by firstly extracting features from captured workpiece images, followed by analysis using machine learning algorithms. This enables inspection to be performed more reliably with minimal human intervention. In this paper, three visual texture features are investigated to classify machined workpiece surfaces into the six machining process classes of Turning, Grinding, Horizontal Milling, Vertical Milling, Lapping, and Shaping. These are the multi-directional Gabor filtered images, intensity histogram and edge features statistics. Support Vector Machines (SVM) applying different kernel functions are investigated for best classifier performance. Results indicate that the Gabor-based SVM-linear kernel provides superior performance.

This paper describes Support Vector Machine (SVM) and Artificial Neural Network (ANN) based recognition and classification of visual symptoms affected by fungal disease. Color images of fungal disease symptoms affected on cereals like wheat, maize and jowar are used in this work. Different types of symptoms affected by fungal disease namely leaf blight, leaf spot, powdery mildew, leaf rust, smut are considered for the study. The developed algorithms are used to preprocess, segment, extract features from disease affected regions. The affected regions are segmented using k-means segmentation technique. Color texture features are extracted from affected regions and then used as inputs to SVM and ANN classifiers. The texture analysis is done using Color Co-occurrence Matrix. Tests are performed to classify image samples. Classification accuracies between 68.5% and 87% are obtained using ANN classifier. The average classification accuracies have increased to 77.5% and 91.16% using SVM classifier.

Flower classification is a useful way for grouping a flower in certain class using specific features. This research propose a new method of flower classification system using combination of color and texture features. The first phase is getting the crown of the flower, which is localized from a flower image by using pillbox filtering and OTSU’s thresholding. In the next phase, color and texture features are extracted from the crown. The color features are extracted by removing L channel in L*a*b* color space, and taking only a* and b* channel, because of ignoring different lighting condition in flower image. The texture features are extracted by Segmentation-based Fractal Texture Analysis (SFTA). The combination features which are consisted of 10 color features and 48 texture features are used as input in k-Nearest Neighbor (kNN) classifier method with cosine distance. The flower classification achieves the best result with accuracy 73.63%.

In this paper, lesion areas affected by anthracnose are segmented using segmentation techniques, graded based on percentage of affected area and neural network classifier is used to classify normal and anthracnose affected on fruits. We have considered three types of fruit namely mango, grape and pomegranate for our work. The developed processing scheme consists of two phases. In the first phase, segmentation techniques namely thresholding, region growing, K-means clustering and watershed are employed for separating anthracnose affected lesion areas from normal area. Then these affected areas are graded by calculating the percentage of affected area. In the second phase texture features are extracted using Runlength Matrix. These features are then used for classification purpose using ANN classifier. We have conducted experimentation on a dataset of 600 fruits’ image samples. The classification accuracies for normal and affected anthracnose fruit types are 84.65% and 76.6% respectively. The work finds application in developing a machine vision system in horticulture field.

본 논문에서는 이미지에 대한 공간 특성(Spatial properties) 및 통계적 특성(Statistical properties)을 포함한 특징이미지를 구성하고, 지역 분산 크기를 이용한 공분산 행렬을 생성하여 텍스쳐 분류에 이용함으로서 조도(illumination) 및 노이즈(Noise) 그리고 회전(Rotation)에 강인한 텍스쳐 분류 방법을 제안한다. 또한 영역 합계의 빠른 연산을 위해 사용된 중간 이미지 표현인 적분 이미지(Integral Image)를 이용함으로서 텍스쳐 검출 프로세스의 수행 시간을 최소화 하는 방법을 제공한다. 제안한 방법의 성능 평가를 위해 브로다츠(Brodatz) 질감 이미지를 이용하여 잡음 추가 및 히스토그램 명세화 그리고 회전 이미지를 생성하여 실험하였으며, 96% 이상의 성능을 얻을 수 있었다.

In this paper, we make up a feature image including spatial properties and statistical properties on image, and format covariance matrices using region variance magnitudes. By using it to texture classification, this paper puts a proposal for tough texture classification way to illumination, noise and rotation. Also we offer a way to minimalize performance time of texture classification using integral image expressing middle image for fast calculation of region sum. To estimate performance evaluation of proposed way, this paper use a Brodatz texture image, and so conduct a noise addition and histogram specification and create rotation image. And then we conduct an experiment and get better performance over 96%

Embedded systems are becoming more popular as many embedded platforms have become more affordable. It offers a compact solution for many different problems including computer vision applications. Texture classification can be used to solve various problems, and implementing it in embedded platforms will help in deploying these applications into the market. This paper proposes to deploy the texture classification algorithms onto the embedded computer vision (ECV) platform. Two algorithms are compared; grey level co-occurrence matrices (GLCM) and Gabor filters. Experimental results show that raw GLCM on MATLAB could achieves 50ms, being the fastest algorithm on the PC platform. Classification speed achieved on PC and ECV platform, in C, is 43ms and 3708ms respectively. Raw GLCM could achieve only 90.86% accuracy compared to the combination feature (GLCM and Gabor filters) at 91.06% accuracy. Overall, evaluating all results in terms of classification speed and accuracy, raw GLCM is more suitable to be implemented onto the ECV platform.

In the process of the sonar image textures produced, the orientation and scale factors are very significant. However, most of the related methods ignore the directional information and scale invariance or just pay attention to one of them. To overcome this problem, we apply Gabor wavelet to extract the features of sonar images, which combine the advantages of both the Gabor filter and traditional wavelet function. The mother wavelet is designed with constrained parameters and the optimal parameters will be selected at each orientation, with the help of bandwidth parameters based on the Fisher criterion. The Gabor wavelet can have the properties of both multi-scale and multi-orientation. Based on our experiment, this method is more appropriate than traditional wavelet or single Gabor filter as it provides the better discrimination of the textures and improves the recognition rate effectively. Meanwhile, comparing with other fusion methods, it can reduce the complexity and improve the calculation efficiency.

This research uses high-resolution satellite images as a source of collecting farmland information. For effectively extract the paddy area, we use texture information and different classify methods to assist the satellite image classification. First, using maximum likelihood classifier to extract paddy information from images. The results show that User Accuracy and Procedure Accuracy of the paddy area can increase from 80.60% to 95.45% and 84.38% to 95.45%. Second, establishing a paddy Knowledge Base and using Knowledge Base Classifier to extract paddy area, and result shows the User Accuracy and Producer Accuracy to be 92.16% and 90.06%. Finally, The result shows we can effectively contribute to the paddy field information extraction from high-resolution satellite images.

텍스쳐 분류에 사용되는 방식 중 하나인 지역적 이진화 패턴은 일반적으로 영상 내의 평탄한 부분, 에지, 코너의 분포를 사용한다. 그러나 영상이 가지는 방향성을 고려하지 않고, 단순히 크고 작음만을 비교하는 지역적 이진화 패턴의 특성때문에 화소간 차이를 반영하지 못하는 문제점이 있다. 또한 영상의 분포를 사용하기 때문에 작은 크기의 영상에 대해서는 분류 성능이 저하된다. 이런 문제를 해결하기 위해 본 논문에서는 영상의 방향성 분포와 고유치 행렬을 이용한 세부 분류 기법을 제안한다. 지역적 이진화 패턴으로 초기 분류에서 누락된 텍스쳐 영상에 대하여 두 가지 특징을 이용하여 세부적으로 분류한다. 첫째, 영상이 가질 수 있는 방향을 여덟 가지로 양자화하고 그 방향들의 분포를 계산한다. 둘째, 구조 행렬을 이용하여 나온 고유치 중 큰 값의 분포를 구한다. 모의 실험을 통해 지역적 이진화 패턴만을 사용하였을 때 대비 제안 방법이 약 8% 정도 분류 정확도가 향상됨을 보였다.

The local binary pattern (LBP) scheme that is one of the texture classification methods normally uses the distribution of flat, edge and corner patterns. However, it cannot examine the edge direction and the pixel difference because it is a sort of binary pattern caused by thresholding. Furthermore, since it cannot consider the pixel distribution, it shows lower performance as the image size becomes larger. In order to solve this problem, we propose a sub-classification method using the edge direction distribution and eigen-matrix. The proposed sub-classification is applied to the particular texture patches which cannot be classified by LBP. First, we quantize the edge direction and compute its distribution. Second, we calculate the distribution of the largest value among eigenvalues derived from structure matrix. Simulation results show that the proposed method provides a higher classification performance of about 8 % than the existing method.

최근 영상 검색(retrieval)과 분류(classification)에서 질감 특징(texture feature)을 이용한 연구들이 활발하게 진행되고 있다. 본 논문에서는 효율적인 질감 특징 추출을 위해 명암도 상호발생 행렬법(gray level co-occurrence matrix)과 웨이블릿 변환(wavelet transform)을 이용하여 질감의 특징을 추출한 후 특징의 중요도에 따라서 가중치를 부여하는 방법을 제안한다. 이렇게 추출된 가중치 대표 벡터들을 기반으로 베이시안 분류기(Bayesian classifier)를 통해 임의의 질감을 인식하였다.

Mathematical morphology based on set theory is easy to be implemented in parallel and can be applied to various fields in image analysis. Particularly mophological pattern spectrum can detect critical scales in an image object and quantify various aspects of the shape-size content. In this paper, texture classification using pattern spectrum based on morphological subband decomposition is porposed. The low-low band extracts pattern spectrum features, and the high-low, low-high, and high-high bands extrack the structural information. This approach has the advantages of efficient information extraction, less time-consuming, high accuacy, less computation, and parallel implementation.