Image Classification by Using Multiclass Support Vector Machines

International Journal of Computer Applications, 2014

Content Based Image Retrieval (CBIR) is a traditional and developing trend in Digital Image Processing. Therefore the use of CBIR to search and retrieve the query image from wide range of database is increasing. In this paper we are going to explore an efficient image retrieval technique which uses local color, shape and texture features. So, efficient image retrieval algorithms based on RGB histograms, Geometric moment and Co-occurrence Model is proposed for color, shape and texture respectively. Results based on this approach are found encouraging in terms of color, shape and texture image classification accuracy. After the features are selected, an SVM classifier is trained to distinguish between relevant and irrelevant images accordingly.

The dramatic rise in the sizes of images databases has stirred the development of effective and efficient retrieval systems. The development of these systems started with retrieving images using textual connotations but later introduced image retrieval based on content. This came to be known as Content Based Image Retrieval or CBIR. Systems using CBIR retrieve images based on visual features such as texture, color and shape, as opposed to depending on image descriptions or textual indexing. In the proposed work we will use various types of image features like color, texture, shape, energy, amplitude and cluster distance to classify the images according to the query image. We will use multi-SVM technique along with clustering technique to compare the features of the input image with the input dataset of images to extract the similar images as that of the query image

2018

Content based image retrieval refers to search for an image from the huge database to retrieve an image with the required shape, text or size. Content-based image retrieval are judged on the parameters such as speed and efficiency, and a modified approach based on a composite color image histogram processing is introduced. This paper introduces a brief study and analysis of Multiclass SVM classifier and comparison of it with KNN classifiers, in application to retrieve a correct output image from database in reference to query image. The proposed approach is time efficient and provides good results. By using a pool of such classifiers we can make a system comprises of multiple CBIR systems providing complementary outputs of each other when yhese outputs are fused together we can have a best match for query image .

In this paper, we propose content-based image retrieval method based on the SVM approach with efficient combination of histogram, color and edge features. Thus extending the work of the previous approach which used same set of features and the Euclidean distance measurement technique. As its histogram features, the extracted histogram bar values for each and every image are used. As its color features, the image is segmented into small pieces and then for each piece the red, green, blue values are used .As its edge features, Canny’s edge detection technique is used to extract the maximum edge value of the image. After extracting the features, we use a machine learning technique called SVM (support vector machine) to find out the optimal result. Combining the features and classifying them using SVM and then finally comparing the results with the previous approach not only gives a better accuracy, but also evaluates the generalization ability under the limited training samples. The analysis of the proposed work is done using MATLAB R2007b simulator.

Foundation of Computer Applications, 2019

The purpose of classification is to analyze the input data and to develop an correct description or model for each class using the features present in the data for its most effective and efficient use. This paper shows comparison of the classifiers for the efficient CBIR system. Image classification is also a machine learning field that uses algorithms mapping all attributes, variables or inputs-function X space-for the definition of class labeled Y. This algorithm is called the classifier. Basically what a classifier does assign a pre-defined class label to a sample. This paper introduces five classifiers (Naïve Bayes, K-Nearest, Artificial Neural Network, Rough Sets and Support Vector Machine). Among them this CBIR system is implemented a support vector machine classification. SVM depend on the concept of the decision plan that determines the boundaries of the decision. SVM classifiers can be learned from relevant and irrelevant user-generated image for training data.There are two major steps in the classification system such as training step and testing step. Training defines criteria based on recognized features.

Expert Systems with Applications, 2012

With the evolution of digital technology, there has been a significant increase in the number of images stored in electronic format. These range from personal collections to medical and scientific images that are currently collected in large databases. Many users and organizations now can acquire large numbers of images and it has been very important to retrieve relevant multimedia resources and to effectively locate matching images in the large databases. In this context, content-based image retrieval systems (CBIR) have become very popular for browsing, searching and retrieving images from a large database of digital images with minimum human intervention. The research community are competing for more efficient and effective methods as CBIR systems may be heavily employed in serving time critical applications in scientific and medical domains. This paper proposes an extremely fast CBIR system which uses Multiple Support Vector Machines Ensemble. We have used Daubechies wavelet transformation for extracting the feature vectors of images. The reported test results are very promising. Using data mining techniques not only improved the efficiency of the CBIR systems, but they also improved the accuracy of the overall process.

Retrieval of an image is a more effective and efficient for managing extensive image database. Content Based Image Retrieval (CBIR) is a one of the image retrieval technique which uses user visual features of an image such as color, shape, and texture features etc. It permits the end user to give a query image in order to retrieve the stored images in database according to their similarity to the query image. In this work, content based image retrieval is accomplished by combining the two features such as color and texture. Color features are extracted by using hsv histogram, color correlogram and color moment values. Texture features are extracted by Segmentation based Fractal Texture Analysis (SFTA). The combined features which are made up of 32 histogram values,64 color correlogram values, 6 color moment values and 48 texture features are extracted to both query and database images. The extracted feature vector of the query image is compared with extracted feature vectors of the database images to obtain the similar images. The main objective this work is classification of image using SVM algorithm.

International Journal of Science Technology & Engineering

The fast growth of computer technologies and the coming of the World Wide Web have increased the amount and the complex difficulty of combining video, sound, words and pictures together information. A Content Based Image Retrieval (CBIR) system has been developed as an efficient image retrieval tool where by the user can provide their question to the system to allow it to retrieve the user’s desired image from the image collection. However the usual relevance responses to something or helpful returned information method to support the user question based on the representative image selection and weight ranking of the images retrieved. The Support Vector Machine(SVM) has been used to support the learning process to reduce the semantic gap between the user and the CBIR system.SVM can classify the data into relevance training set and Gabor Filtering will extract the feature from the given image dataset. It can also improve the performance of CBIR. Also solve the imbalance training set.

As the growth and development of various multimedia technologies in the field of CBIR many advanced information retrieval systems have become popular and has brought the new evolution in fast and effective retrieval. In this paper the techniques of image classification in CBIR are been discussed and compared. It also introduces classifiers like support vector machine, Bayesian classifier for accurate and efficient retrieval of images.

Content Based Image Retrieval (CBIR) is an image retrieval method that is based on the meta content of an image which is proven to be more efficient than image retrieval method that is based on text or key words. Bag of Visual Words (BoVW) is a popular approach and the most widely used in CBIR problem. In this research, a CBIR method using BoVW and multiclass SVM classifier will be proposed. In BoVW, Scale Invariant Feature Transform (SIFT) will be used as the local features descriptor. This research will use Gaussian Mixture Model (GMM) as the method to do the visual vocabulary generation and Fisher Vector (FV) to create the encoder. The multiclass SVM classifier will use linear kernel, Hellinger's kernel, and chi-square kernel to do the classification. After the classification, the proposed CBIR will use the classification model to classify query image. After the query image class is known, then the color histogram features will be extracted from query image and dataset which only consists of image in the same class as query image. Datasets used in the research are Corel and Guang-Hai Liu (GHIM-10K). The experimental results show that BoVW with GMM and FV produces higher accuracy than other encoding methods in BoVW which produce a good retrieval result in the CBIR. 1. Introduction. The rapid development of digital technology has caused the amount of image data to increase at a rapid pace almost equal to the number of text data. However, searching data in image data is not as easy as searching data in text data. Searching in image (image retrieval) becomes one of the most commonly researched topics because of the need of an effective method to do searching in image data [1]. In general image retrieval is divided into 2 methods, which are text based and content based. Recent research shows that text-based method is not effective to represent an image into a small set of keywords which then shifts recent research topics to content-based method [2]. The idea of Content-Based Image Retrieval (CBIR) is searching an image using Meta data from the image. CBIR uses low level features in the searching method such as color, texture, or shape [3]. In general, the features which CBIR uses can be categorized into 2 key features: global feature descriptors and local feature descriptors. Global feature descriptors describe an image as a whole, while local feature descriptors describe an image by image patch (a small pixel). Global feature descriptors have an advantage in terms of its computational speed but the drawback is the low accuracy. Global feature descriptors tend to fail in identifying important visual characteristics from an image. On the other hand, local feature descriptors produce better accuracy than the global feature descriptors mainly because the image is represented by features calculated from the patch in the image. The drawback of the local features is that the size of feature space is large and will become very huge for large image database [4].