Content-based image retrieval based on supervised learning and statistical-based moments

2019 ◽  
Vol 33 (19) ◽  
pp. 1950213 ◽  
Author(s):  
Vibhav Prakash Singh ◽  
Rajeev Srivastava ◽  
Yadunath Pathak ◽  
Shailendra Tiwari ◽  
Kuldeep Kaur
Keyword(s):  
Image Retrieval ◽  
Supervised Learning ◽  
Texture Features ◽  
Ground Truth ◽  
Local Binary Patterns ◽  
Content Based Image Retrieval ◽  
Support Vector ◽  
Svm Classifier ◽  
Learning Approaches ◽  
Query Image

Content-based image retrieval (CBIR) system generally retrieves images based on the matching of the query image from all the images of the database. This exhaustive matching and searching slow down the image retrieval process. In this paper, a fast and effective CBIR system is proposed which uses supervised learning-based image management and retrieval techniques. It utilizes machine learning approaches as a prior step for speeding up image retrieval in the large database. For the implementation of this, first, we extract statistical moments and the orthogonal-combination of local binary patterns (OC-LBP)-based computationally light weighted color and texture features. Further, using some ground truth annotation of images, we have trained the multi-class support vector machine (SVM) classifier. This classifier works as a manager and categorizes the remaining images into different libraries. However, at the query time, the same features are extracted and fed to the SVM classifier. SVM detects the class of query and searching is narrowed down to the corresponding library. This supervised model with weighted Euclidean Distance (ED) filters out maximum irrelevant images and speeds up the searching time. This work is evaluated and compared with the conventional model of the CBIR system on two benchmark databases, and it is found that the proposed work is significantly encouraging in terms of retrieval accuracy and response time for the same set of used features.

scholarly journals Implementation of early and late fusion methods for content-based image retrieval

2021 ◽  
Vol 8 (7) ◽  
pp. 97-105
Author(s):  
Ali Ahmed ◽  
◽  
Sara Mohamed ◽  
Keyword(s):  
Image Retrieval ◽  
Texture Features ◽  
Image Features ◽  
Distance Measures ◽  
Content Based Image Retrieval ◽  
Late Fusion ◽  
Query Image ◽  
Fusion Methods ◽  
And Performance ◽  
Image Repository

Content-Based Image Retrieval (CBIR) systems retrieve images from the image repository or database in which they are visually similar to the query image. CBIR plays an important role in various fields such as medical diagnosis, crime prevention, web-based searching, and architecture. CBIR consists mainly of two stages: The first is the extraction of features and the second is the matching of similarities. There are several ways to improve the efficiency and performance of CBIR, such as segmentation, relevance feedback, expansion of queries, and fusion-based methods. The literature has suggested several methods for combining and fusing various image descriptors. In general, fusion strategies are typically divided into two groups, namely early and late fusion strategies. Early fusion is the combination of image features from more than one descriptor into a single vector before the similarity computation, while late fusion refers either to the combination of outputs produced by various retrieval systems or to the combination of different rankings of similarity. In this study, a group of color and texture features is proposed to be used for both methods of fusion strategies. Firstly, an early combination of eighteen color features and twelve texture features are combined into a single vector representation and secondly, the late fusion of three of the most common distance measures are used in the late fusion stage. Our experimental results on two common image datasets show that our proposed method has good performance retrieval results compared to the traditional way of using single features descriptor and also has an acceptable retrieval performance compared to some of the state-of-the-art methods. The overall accuracy of our proposed method is 60.6% and 39.07% for Corel-1K and GHIM-10K ‎datasets, respectively.

Medical Image Retrieval Based on Semi-Supervised Learning

2010 ◽  
Vol 108-111 ◽  
pp. 201-206 ◽  
Author(s):  
Hui Liu ◽  
Cai Ming Zhang ◽  
Hua Han
Keyword(s):  
Support Vector Machine ◽  
Active Learning ◽  
Image Retrieval ◽  
Supervised Learning ◽  
Relevance Feedback ◽  
Medical Image ◽  
Content Based Image Retrieval ◽  
Support Vector ◽  
Learning Support ◽  
Medical Image Retrieval

Among various content-based image retrieval (CBIR) methods based on active learning, support vector machine(SVM) active learning is popular for its application to relevance feedback in CBIR. However, the regular SVM active learning has two main drawbacks when used for relevance feedback. Furthermore, it’s difficult to collect vast amounts of labeled data and easy for unlabeled data to image examples. Therefore, it is necessary to define conditions to utilize the unlabeled examples enough. This paper presented a method of medical images retrieval about semi-supervised learning based on SVM for relevance feedback in CBIR. This paper also introduced an algorithm about defining two learners, both learners are re-trained after every relevance feedback round, and then each of them gives every image in a rank. Experiments show that using semi-supervised learning idea in CBIR is beneficial, and the proposed method achieves better performance than some existing methods.

scholarly journals Content Based Image Retrieval using Multi-level 3D Color Texture and Low Level Color Features with Neural Network Based Classification System

2021 ◽  
Vol 15 ◽  
pp. 265-270
Author(s):  
Priyesh Tiwari ◽  
Shivendra Nath Sharan ◽  
Kulwant Singh ◽  
Suraj Kamya
Keyword(s):  
Neural Network ◽  
Image Retrieval ◽  
Texture Feature ◽  
Texture Features ◽  
Classification Model ◽  
Content Based Image Retrieval ◽  
Query Image ◽  
Multi Level ◽  
Different Color ◽  
Color Texture

Content based image retrieval (CBIR), is an application of real-world computer vision domain where from a query image, similar images are searched from the database. The research presented in this paper aims to find out best features and classification model for optimum results for CBIR system.Five different set of feature combinations in two different color domains (i.e., RGB & HSV) are compared and evaluated using Neural Network Classifier, where best results obtained are 88.2% in terms of classifier accuracy. Color moments feature used comprises of: Mean, Standard Deviation,Kurtosis and Skewness. Histogram features is calculated via 10 probability bins. Wang-1k dataset is used to evaluate the CBIR system performance for image retrieval.Research concludes that integrated multi-level 3D color-texture feature yields most accurate results and also performs better in comparison to individually computed color and texture features.

scholarly journals An Efficient Two-layer based Technique for Content-based Image Retrieval

2021 ◽  
Vol 5 (1) ◽  
pp. 28
Author(s):  
Fawzi Abdul Azeez Salih ◽  
Alan Anwer Abdulla
Keyword(s):  
Image Retrieval ◽  
Texture Features ◽  
Superior Performance ◽  
Content Based Image Retrieval ◽  
Discrete Wavelet ◽  
Color Spaces ◽  
Query Image ◽  
Current State ◽  
Similar Images ◽  
Different Color

The rapid advancement and exponential evolution in the multimedia applications raised the attentional research on content-based image retrieval (CBIR). The technique has a significant role for searching and finding similar images to the query image through extracting the visual features. In this paper, an approach of two layers of search has been developed which is known as two-layer based CBIR. The first layer is concerned with comparing the query image to all images in the dataset depending on extracting the local feature using bag of features (BoF) mechanism which leads to retrieve certain most similar images to the query image. In other words, first step aims to eliminate the most dissimilar images to the query image to reduce the range of search in the dataset of images. In the second layer, the query image is compared to the images obtained in the first layer based on extracting the (texture and color)-based features. The Discrete Wavelet Transform (DWT) and Local Binary Pattern (LBP) were used as texture features. However, for the color features, three different color spaces were used, namely RGB, HSV, and YCbCr. The color spaces are utilized by calculating the mean and entropy for each channel separately. Corel-1K was used for evaluating the proposed approach. The experimental results prove the superior performance of the proposed concept of two-layer over the current state-of-the-art techniques in terms of precision rate in which achieved 82.15% and 77.27% for the top-10 and top-20, respectively.

Classification of Crop Residue Cover in High-Resolution RGB Images Using Machine Learning

2022 ◽  
Vol 65 (1) ◽  
pp. 75-86
Author(s):  
Parth C. Upadhyay ◽  
John A. Lory ◽  
Guilherme N. DeSouza ◽  
Timotius A. P. Lagaunne ◽  
Christine M. Spinka
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Feature Selection Method ◽  
Texture Features ◽  
Ground Truth ◽  
Selection Method ◽  
Support Vector ◽  
Svm Classifier ◽  
Aerial Vehicle ◽  
Rgb Images

HighlightsA machine learning framework estimated residue cover in RGB images taken at three resolutions from 88 locations.The best results primarily used texture features, the RFE-SVM feature selection method, and the SVM classifier.Accounting for shadows and plants plus modifying and optimizing the texture features may improve performance.An automated system developed using machine learning is a viable strategy to estimate residue cover from RGB images obtained with handheld or UAV platforms.Abstract. Maintaining plant residue on the soil surface contributes to sustainable cultivation of arable land. Applying machine learning methods to RGB images of residue could overcome the subjectivity of manual methods. The objectives of this study were to use supervised machine learning while identifying the best feature selection method, the best classifier, and the most effective image feature types for classifying residue levels in RGB imagery. Imagery was collected from 88 locations in 40 row-crop fields in five Missouri counties between early May and late June in 2018 and 2019 using a tripod-mounted camera (0.014 cm pixel-1 ground sampling distance, GSD) and an unmanned aerial vehicle (UAV, 0.05 and 0.14 GSD). At each field location, 50 contiguous 0.3 × 0.2 m region of interest (ROI) images were extracted from the imagery, resulting in a dataset of 4,400 ROI images at each GSD. Residue percentages for ground truth were estimated using a bullseye grid method (n = 100 points) based on the 0.014 GSD images. Representative color, texture, and shape features were extracted and evaluated using four feature selection methods and two classifiers. Recursive feature elimination using support vector machine (RFE-SVM) was the best feature selection method, and the SVM classifier performed best for classifying the amount of residue as a three-class problem. The best features for this application were associated with texture, with local binary pattern (LBP) features being the most prevalent for all three GSDs. Shape features were irrelevant. The three residue classes were correctly identified with 88%, 84%, and 81% 10-fold cross-validation scores for the 2018 training data and 81%, 69%, and 65% accuracy for the 2019 testing data in decreasing resolution order. Converting image-wise data (0.014 GSD) to location residue estimates using a Bayesian model showed good agreement with the location-based ground truth (r2 = 0.90). This initial assessment documents the use of RGB images to match other methods of estimating residue, with potential to replace or be used as a quality control for line-transect assessments. Keywords: Feature selection, Soil erosion, Support vector machine, Texture features, Unmanned aerial vehicle.

Content-based Image Retrieval by Information Theoretic Measure

2011 ◽  
Vol 61 (5) ◽  
pp. 415 ◽  
Author(s):  
Madasu Hanmandlu ◽  
Anirban Das
Keyword(s):  
Image Retrieval ◽  
Feature Vector ◽  
Texture Features ◽  
Entropy Function ◽  
Content Based Image Retrieval ◽  
Query Image ◽  
Information Theoretic ◽  
Dominant Colour ◽  
Image Retrieval System ◽  
Information Theoretic Measures

<p>Content-based image retrieval focuses on intuitive and efficient methods for retrieving images from databases based on the content of the images. A new entropy function that serves as a measure of information content in an image termed as 'an information theoretic measure' is devised in this paper. Among the various query paradigms, 'query by example' (QBE) is adopted to set a query image for retrieval from a large image database. In this paper, colour and texture features are extracted using the new entropy function and the dominant colour is considered as a visual feature for a particular set of images. Thus colour and texture features constitute the two-dimensional feature vector for indexing the images. The low dimensionality of the feature vector speeds up the atomic query. Indices in a large database system help retrieve the images relevant to the query image without looking at every image in the database. The entropy values of colour and texture and the dominant colour are considered for measuring the similarity. The utility of the proposed image retrieval system based on the information theoretic measures is demonstrated on a benchmark dataset.</p><p><strong>Defence Science Journal, 2011, 61(5), pp.415-430</strong><strong><strong>, DOI:http://dx.doi.org/10.14429/dsj.61.1177</strong></strong></p>

scholarly journals Proposed a Content-Based Image Retrieval System Based on the Shape and Texture Features

2019 ◽  
Vol 8 (11) ◽  
pp. 2185-2192
Keyword(s):  
Image Retrieval ◽  
Texture Feature ◽  
Texture Features ◽  
Visual Image ◽  
Content Based Image Retrieval ◽  
Query Image ◽  
Feature Extraction Method ◽  
Image Retrieval System ◽  
Reference Images ◽  
Similar Images

In this paper, we proposed a fusion feature extraction method for content based image retrieval. The feature is extracted by focusing on the texture and shape features of the visual image by using the Local Binary Pattern (LBP – texture feature) and Edge Histogram Descriptor (EHD – shape feature). The SVD is used for decreasing the number of the feature vector of images. The Kd-tree is used for reducing the retrieval time. The input to this system is a query image and Database (the reference images) and the output is the top n most similar images for the query image. The proposed system is evaluated by using (precision and recall) to measure the retrieval effectiveness. The values of the recall are between [43% –93%] and the average recall is 64.3%. The values of precision are between [30%-100%] and the average is 72.86% for the entire system and for both databases

scholarly journals Knn And Steerable Pyramid Based Enhanced Content Based Image Retrieval Mechanism

2018 ◽  
Vol 17 (2) ◽  
pp. 7215-7225
Author(s):  
Bohar Singh ◽  
Mrs. Mehak Aggarwal
Keyword(s):  
Image Retrieval ◽  
Image Data ◽  
Texture Features ◽  
Content Management ◽  
Content Based Image Retrieval ◽  
Visual Content ◽  
Image Region ◽  
Query Image ◽  
Image Retrieval System ◽  
Steerable Pyramid

Recently, digital content has become a significant and inevitable asset of or any enterprise and the need for visual content management is on the rise as well. There has been an increase in attention towards the automated management and retrieval of digital images owing to the drastic development in the number and size of image databases. A significant and increasingly popular approach that aids in the retrieval of image data from a huge collection is called Content-based image retrieval (CBIR). Content-based image retrieval has attracted voluminous research in the last decade paving way for development of numerous techniques and systems besides creating interest on fields that support these systems. CBIR indexes the images based on the features obtained from visual content so as to facilitate speedy retrieval. Content based image retrieval from large resources has become an area of wide interest nowadays in many applications. In this thesis work, we present a steerable pyramid based image retrieval system that uses color, contours and texture as visual features to describe the content of an image region. To speed up retrieval and similarity computation, the database images are classified and the extracted regions are clustered according to their feature vectors using KNN algorithm We have used steerable pyramid to extract texture features from query image and classified database images and store them in feature features. Therefore to answer a query our system does not need to search the entire database images; instead just a number of candidate images are required to be searched for image similarity.  Our proposed system has the advantage of increasing the retrieval accuracy and decreasing the retrieval time.

scholarly journals CONTENT BASED IMAGE RETRIEVAL

2013 ◽  
pp. 294-299
Author(s):  
HARSHADA ANAND KHUTWAD ◽  
RAVINDRA JINADATTA VAIDYA
Keyword(s):  
Image Retrieval ◽  
Wavelet Decomposition ◽  
Texture Features ◽  
Content Based Image Retrieval ◽  
Image Search ◽  
Query Image ◽  
Shape Information ◽  
Color Correlogram ◽  
Similar Images ◽  
The Given

Content Based Image Retrieval is an interesting and most emerging field in the area of ‘Image Search’, finding similar images for the given query image from the image database. Current approaches include the use of color, texture and shape information. Considering these features in individual, most of the retrievals are poor in results and sometimes we are getting some non relevant images for the given query image. So, this dissertation proposes a method in which combination of color and texture features of the image is used to improve the retrieval results in terms of its accuracy. For color, color histogram based color correlogram technique and for texture wavelet decomposition technique is used. Color and texture based image

scholarly journals Framework for Color and Texture Feature Fusion in Content Based Image Retrieval using Block Truncation Coding with Color Spaces

2020 ◽  
Vol 9 (3) ◽  
pp. 769-774
Keyword(s):  
Image Retrieval ◽  
Feature Fusion ◽  
Texture Feature ◽  
Local Binary Patterns ◽  
Input Image ◽  
Content Based Image Retrieval ◽  
Color Spaces ◽  
Query Image ◽  
Block Truncation Coding ◽  
Rate Of Increase

With tremendous growth in social media and digital technologies, generation, storing and transfer of huge amount of information over the internet is on the rise. Images or visual mode of communication have been prevailing and widely accepted as a mode of communication since ages. And with the growth of internet, the rate at which images are generated is growing exponentially. But the methods used to retrieve images are still very slow and inefficient, compared to the rate of increase in image databases. To cope up with this explosive increase in images, this information age has seen huge research advancement in Content Based Image Retrieval (CBIR). CBIR systems provide a way of utilizing the 3 major ways in which content is portrayed in images, those are shape, texture and color. In CBIR system, features are extracted from query image and similarity is found with features stored in database for retrieval. This provides an objective way of image retrieval, which is more efficient compared to subjective human annotation. Application specific CBIR systems have been developed and perform really well, but Generic CBIR systems are still under developed. Block Truncation Coding (BTC) has been chosen as a feature extractor. BTC applied directly on input image provides color content-based features of image and BTC applied after applying LBP on the image provide texture content-based features of image. Previous work consists of either color, shape or texture, but usage of more than one descriptor is still in research and might give better performance. The paper presents framework for color and texture feature fusion in content-based image retrieval using block truncation coding with color spaces. Experimentation is carried out on Wang Dataset of 1000 images consisting of 10 classes. Each class has 100 images in it. Obtained results have shown performance improvement using fusion of BTC extracted color features and texture features extracted with BTC applied on Local Binary Patterns (LBP). Conversion of color space from RGB to LUV is done using Kekre's LUV.

Sign in / Sign up

Export Citation Format

Share Document