Texture feature extraction using gray level statistical matrix for content-based mammogram retrieval

Chandy, D. Abraham; Johnson, Jeffrey S.; Selvan, S.

doi:10.1007/s11042-013-1511-z

Cited by 32 publications

(15 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All different cases tested in the following points are shown in Table I. First of all, when the Naive Bayes classifier is trained with 40% of the original textures, using all 104 features proposed in section 2 and section 3, without any filter nor PCA process, the mean success of the classifier in a total of 10,000 tests 6 is 72.28%, with a maximum value of 78.60% and a standard deviation of 1.97%. This result, for all the tested permutations, can be seen in Fig.…”

Section: Classification Resultsmentioning

confidence: 99%

“…Now, assuming that 3-gray levels 2 -which here have been called "A", "B" and "C"-are defined a new image I K can be described by (6). Thus, the original image can be rewritten, considering (6), as shown in (7).…”

Section: Haralick Texture Featuresmentioning

confidence: 99%

“…The extraction of texture information has shown positive results when used for characterizing and estimate parameters of any image [1], [2], [3], [4], [5], [6] and [7]. Each parameter will depend strongly on the type of image to be processed (i.e.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Texture Classification based on Spectral Analysis and Haralick Features

Valentin

Albuquerque

et al. 2016

View full text Add to dashboard Cite

In this work we discuss a method to classify a set of texturized images based on the extraction of their Haralick Features. This kind of Classification is capable of providing texture-based measurements (such as contrast or correlation) and use them as main parameters to classify the same type of patterns in other images. In order to improve the classification success ratio a spectral analysis of the textures and, therefore, the use of filters, before the classification step, is proposed here. In this work the classification success has been evaluated using Mean and Canny filters. On the other hand, the Principal Component Analysis is used to optimize the features extracted for the patterns on each image, before introduced into the classifier. With this method the classifying success ratio for the KTH-TIPS subset and 10,000 different permutations was increased-in averagefrom 72.28% to 84.25%.

show abstract

Section: Classification Resultsmentioning

confidence: 99%

Section: Haralick Texture Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Texture Classification based on Spectral Analysis and Haralick Features

Valentin

Albuquerque

et al. 2016

View full text Add to dashboard Cite

show abstract

“…The GLCM is the conventional way of extracting statistical texture features [39]. It works by forming a moving window through the image and then calculating the frequency of the co-occurrence of the pixel values in a defined number of directions.…”

Section: Multiple Feature Extraction and Initial Training Samples Acqmentioning

confidence: 99%

Object-Based Change Detection Using Multiple Classifiers and Multi-Scale Uncertainty Analysis

et al. 2019

View full text Add to dashboard Cite

The drawback of pixel-based change detection is that it neglects the spatial correlation with neighboring pixels and has a high commission ratio. In contrast, object-based change detection (OBCD) depends on the accuracy of the segmentation scale, which is of great significance in image analysis. Accordingly, an object-based approach for automatic change detection using multiple classifiers and multi-scale uncertainty analysis (OB-MMUA) in high-resolution (HR) remote sensing images is proposed in this paper. In this algorithm, the gray-level co-occurrence matrix (GLCM), morphological, and Gabor filter texture features are extracted to construct the input data, along with the spectral features, to utilize the respective advantages of the features and to compensate for the insufficient spectral information. In addition, random forest is used to select the features and determine the optimal feature vectors for the change detection. Change vector analysis (CVA) based on uncertainty analysis is then implemented to select the initial training samples. According to the diversity, support vector machine (SVM), k-nearest neighbor (KNN), and extra-trees (ExT) classifiers are then chosen as the base classifiers for Dempster-Shafer (D-S) evidence theory fusion, and unlabeled samples are selected using an active learning method with spatial information. Finally, multi-scale object-based D-S evidence theory fusion and uncertainty analysis is used to classify the difference image. To validate the proposed approach, we conducted experiments using multispectral images collected by the ZY-3 and GF-2 satellites. The experimental results confirmed the effectiveness and superiority of the proposed approach, which integrates the respective advantages of the pixel-based and object-based methods.Remote Sens. 2019, 11, 359 2 of 17 fields, such as urban development [6], environmental monitoring, vegetation coverage studies [7], and land-use monitoring [8,9].Change detection can be divided into pixel-based and object-based change detection (OBCD). Many scholars have proposed a variety of pixel-based change detection methods, including change vector analysis (CVA) [10,11], Markov random field (MRF)-based change detection [12], etc. With the development of machine learning, methods such as the extreme learning machine (ELM) algorithm [13], support vector machine (SVM) [14], random forest (RF) [15], k-nearest neighbor (KNN) [16], the multi-layer perceptron neural network (MLPNN) [17], and convolutional neural network (CNN) [18,19] have improved the accuracy of classification and change detection. However, a single classifier cannot detect all the change information in an image effectively. To address this issue, ensemble learning has been applied to the research and application of change detection and classification [20][21][22]. [23] proposed an ensemble system based on multiple classifiers, and achieved good classification results. Zhang et al. (2017) [24] combined deep learning with feature change analysis for remote sensing image change d...

show abstract

“…Texture is a spatial distribution of the pixels in the image, which can be expressed by the correlation between neighboring pixels [27,28]. Texture analysis is a process of a qualitative or quantitative description of the texture that is extracted by image processing technology.…”

Section: Texture Feature Extractionmentioning

confidence: 99%

Automatic classification of insulator by combining k-nearest neighbor algorithm with multi-type feature for the Internet of Things

Zhu

Huang

et al. 2018

J Wireless Com Network

View full text Add to dashboard Cite

New algorithms and architectures in the context of 5G shall be explored to ensure the efficiency, robustness, and consistency in variable application environments which concern different issues, such as the smart grid, water supply, gas monitoring, etc. In power line monitoring, we can get lots of images through a wide range of sensors and can ensure the safe operation of the smart grid by analyzing images. Feature extraction is critical to identify insulators in the aerial image. Existing approaches have primarily addressed this problem by using a single type of feature such as color feature, texture feature, or shape feature. However, a single type of feature usually leads to poor classification rates and missed detection in identifying insulators. Aiming to fully describe the characteristics of insulator and enhance the robustness of insulator against the complex background in aerial images, we combine three types of feature including color feature, texture feature, and shape feature towards a multi-type feature. Then, the multi-type feature is integrated with k-nearest neighbor classifier for automatic classifying insulators. Our experiment with 4500 aerial images demonstrates that the recognition rate is 99% by using this multi-type feature. Comparing to a single type of feature, our method yielded a better classification performance.

show abstract

Texture feature extraction using gray level statistical matrix for content-based mammogram retrieval

Cited by 32 publications

References 20 publications

Texture Classification based on Spectral Analysis and Haralick Features

Texture Classification based on Spectral Analysis and Haralick Features

Object-Based Change Detection Using Multiple Classifiers and Multi-Scale Uncertainty Analysis

Automatic classification of insulator by combining k-nearest neighbor algorithm with multi-type feature for the Internet of Things

Contact Info

Product

Resources

About