Several CBIR schemes have been devised by the formation of the low dimensional image feature vector from the primitive visual features such as color, texture and/or shape of the image to speed up the retrieval process. In this paper, the visual contents of the images have been extracted using block level Discrete Cosine Transformation (DCT) and Gray Level Co-occurrence Matrix (GLCM). Since the DC coefficients based feature vector has retained the most significant visual components of the image, so initially, we have computed DC coefficients based uniform quantized histogram and some statistical parameters are derived from that histogram for the formation of the DC feature vector. Subsequently, other significant visual features are computed from the residual image where the residual image is obtained by taking the difference between the original image plane and the DC coefficients based reconstructed image plane. Thereafter, some statistical parameters from GLCMs of the residual image are considered for the construction of the GLCM based feature vector. This feature vector is suitable to identify the texture features of the residual image in a more effective way. The single feature vector has been obtained by combining DC and GLCM feature vectors since the combined extracted features from the images increase the accuracy of any image retrieval system. We have tested the scheme either in intensity image or on three color planes of an RGB color image. The experimental results are evaluated on two standard image databases and demonstrate the effectiveness of the proposed scheme in terms of average retrieval accuracy. In addition, the overall speed of the proposed CBIR system is high due to the formation of the low dimensional significant feature vector. The comparative results also show that the proposed scheme provides effective accuracy than some other state-of-the-art CBIR schemes.

A novel image retrieval scheme using gray level co-occurrence matrix descriptors of discrete cosine transform based residual image

The modern world uses digital data invariably and the advent of smart phones and cameras add to images and communicating through images in social media is a very common happening due to the exposure to internet and the economical availability of data. Moreover, people explore the web for much news and on different topics. They feel satisfied to look into image type info than textual information. So, image retrieval is paramount nowadays and it is very much useful in many societal and defence applications. The methods used for retrieval started through text based and query based is popularly used now. The content based retrieval uses many stages, approaches, technologies, algorithms etc. It is intended to provide a broad perspective about CBIR in this work. Moreover, the computational and time complexities involved during retrieval relies on the performance of the entire system. One of the aspects that can improve the precision is the reduction of semantic gap. Taking this as the base, we have explored a literature and presented the points related to various points including relevance feedback so as to enable other researchers to carry out experimental work on the suggested areas.

Essentiality for bridging the gap between low and semantic level features in image retrieval systems: an overview

This article proposes stochastic exploration algorithms for mobile robot exploration problems. Navigation with uncertain conditions in the absence of initial parameters is a situation wherein precomputation and prediction are impossible for a robot. Therefore, stochastic optimization techniques were applied to find the optimal solution for the robot exploration problem. Driving to the unknown areas, the robot updates the frontier line of sensor visibility during the exploration mission. The points of the frontier line are assumed as the swarm population with their own positions and costs, which allows the computation of the next global waypoint. The calculation of global waypoints is carried out by a nature-inspired optimization algorithm that can place a waypoint in uncertainties. This study offers to apply three metaheuristic algorithms individually, such as Whale Optimization, Grey Wolf Optimizer, and Particle Swarm Optimization algorithms, for comparison and testing their performances in the mobile robotics. At first, the simulations based on the proposed exploration algorithms were implemented and evaluated in a created environment. The results were compared in a single and average cases. Then, the real-world experiments using Grey Wolf Optimizer exploration algorithm were conducted in the different types of environments using MATLAB-ROS integration tool. These results proved the effectiveness and applicability of the bio-inspired optimization algorithm in the mobile robotics.

/pdf/waypoint-mobile-robot-exploration-based-on-biologically-1hjol5azpy.pdf

Waypoint Mobile Robot Exploration Based on Biologically Inspired Algorithms

A Novel Similarity Measure for Content Based Image Retrieval in Discrete Cosine Transform Domain

We discuss the development of a structural descriptor for object detection in images. The descriptor is based on a graph, whose vertices are the centers of mass of segment features.  The embedding of the graph in a vector space is implemented using a Young-Householder decomposition and based on differential geometry. Compound curves are used to describe the relationship between the points. The image graph is described by a matrix of curvature parameters. The distance matrix for the graphs of the candidate object and the reference object is calculated using the Hausdorff metric. A multidimensional scaling method is used to represent the results. Images of test objects and images of human faces are used to study the developed approach. A comparison of the developed descriptor with the Viola-Jones method is performed when detecting a human head in the image. The advantage of the developed approach is the image rotational invariance in the plane while searching for objects. The descriptor can detect objects rotated in space by angles of up to 50 degrees. Using the mass centers of segments of features as the graph vertices makes the approach more robust to changes in image acquisition angles in comparison with the approach that uses image features as the graph vertices.

/pdf/object-detection-in-images-with-a-structural-descriptor-40sdnvayye.pdf

Object detection in images with a structural descriptor based on graphs

This paper is presents a Comprehensive Survey on Tabu Search Algorithms (TSA). TSA is a meta heuristics kind of algorithm which works on global optimal solution for a given problem such as vehicle routing problem (VRP), open vehicle routing problem (OVRP), multi-trip vehicle routing and scheduling problem (MTVRSP), container loading problem (CLP) and job shop problem, etc. in this paper focuses on main characteristics of TSA and its behaviour.

Tabu Search Algorithm (TSA): A Comprehensive Survey

In this paper we discuss all the CBIR technique and analysis the methods. We also compare all the feature extraction techniques in the tabular form. Color features are extracted with the help of color moments, color histogram, invariant color histogram, and dominant color. Color based CBIR technique there are limitations in the mention method in the table. So we decide to enhance the color based method in the content based image retrieval method. We analysis the all color based method which is used in the proposed method by authors. But color classification methods are not discussed in the above methods. So in the CBIR system if we used the color classification method based on the red, green and blue channel. We easily get the object as well as color. And we also studied various texture and shape based technique that is used by many authors in our research. Texture based various techniques are Gray Level Cooccurrence matrix (GLCM), Gabor Transform and Tamura Features. Texture descriptor provides a measure of properties such as smoothness, roughness, and regularity. The texture of the region is structural, statistical and spectral are three principal approaches used in image processing. Shape features are extracted using many approaches like as Histogram of Edge Directions, Region Moments, invariant Moments, Zernike moments, Legendre Moments.

A Survey on CBIR on the Basis of Different Feature Descriptor

Nowadays on the internet, lots of information is spread every second by the people. On social media, most of the users do not verify the information and propagate it. Manually identifying fake news is a very tremendous problem for all. So, the need for an automatic system that efficiently detects fake news. This paper estimated a model that intuitionally distinguishes fake news from a news article. A new feature set for machine learning classifier has been proposed. Within the experiment, the dataset used has a combination of two datasets that contain equal true news and fake news articles of politics. From text fields of the dataset extract linguistic/stylometric features, a bag of words TF and BOW TF-IDF vector, after that apply the various machine learning models including bagging and boosting methods to achieve the best accuracy.

Machine Learning based Fake News Detection using linguistic features and word vector features

Now a day’s development of multimedia technology, the possibilities of utility of large databases is rapidly increasing. To handle its management and retrieval CBIR is the best and effective method. CBIR technique uses the visual contents like as color, shape and texture that are called features, to searching, browsing, and navigation of query images for large image databases. Color is the visual perceptual property corresponding in humans to the categories called red, blue and yellow etc. Texture is the image and especially physical quality of a surface. Texture is the characteristic structure of the interwoven or intertwined outfit, strands or the like that make up a textile fabric. In this paper we present utility of CBIR system with color and texture features. And we design a color filter with the help of extract the red channel, green channel and blue channel from the original image. After it we find a texture of all channels with the help of statistical method. The combination of texture of red channel, green channel and blue channel we create a feature vector for all images of the database. Experimental results are shows the average accuracy, average precision rate and average retrieval rate. That is better than other existing method.

/pdf/an-efficient-technique-for-image-retrieval-from-the-large-33qgrhstvx.pdf

Mayank Jain

Papers

Tabu Search Algorithm (TSA): A Comprehensive Survey

A Survey on CBIR on the Basis of Different Feature Descriptor

Machine Learning based Fake News Detection using linguistic features and word vector features

An Efficient Technique for Image Retrieval from the Large Database on the Basis of Color and Texture