image retrieval techniques finds very useful in content management systems. Content Based image Retrieval (CBIR) methods incorporates specific image features such as colors, textures, keypoints etc., for retrieving si...
详细信息
ISBN:
(纸本)9781467385640
image retrieval techniques finds very useful in content management systems. Content Based image Retrieval (CBIR) methods incorporates specific image features such as colors, textures, keypoints etc., for retrieving similar images from databases. Most of the keypoint detectors and keypoint descriptors uses only grayscale information. The retrieval accuracy of these methods can be improved by adding additional color information to the keypoint descriptors. In this paper, an enhanced SURF descriptor is proposed for CBIR applications, which extracts the image features by computing the Hu moments along with eigen values in the immediate neighbourhood of the detected keypoints. Experimental results shows better image retrieval accuracy by using enhanced SURF descriptor. Also, the enhanced SURF descriptor is able to differentiate between images of same object having similar grayscale properties but having different colors.
Microaneurysms are small red dots that occur on the retina during preliminary stage of Diabetic Retinopathy. computer aided microaneurysm screening is necessary to prevent the aggravation of the disease and further vi...
详细信息
ISBN:
(纸本)9781467385640
Microaneurysms are small red dots that occur on the retina during preliminary stage of Diabetic Retinopathy. computer aided microaneurysm screening is necessary to prevent the aggravation of the disease and further vision loss. In this paper, Shannon and Tsallis entropy thresholding in conjunction with Naive Bayes classifier is suggested for microaneurysm detection. Various shape and intensity based features are extracted to eliminate the falsely detected candidates. The proposed method is evaluated by plotting the FROC curves using the Retinopathy Online Challenge (ROC) and DIARETDB1 databases. The proposed method achieves high sensitivity values of 0.421 and 0.477 (at false positive rate of 8) using Shannon and Tsallis entropy thresholding which is better than some existing methods.
In recent times, there has been a sharp increase in dengue and malaria, especially in urban areas. One of the major reasons for this health hazard is the number of locations where one can find stagnant water. These lo...
详细信息
ISBN:
(纸本)9781467385640
In recent times, there has been a sharp increase in dengue and malaria, especially in urban areas. One of the major reasons for this health hazard is the number of locations where one can find stagnant water. These locations are large breeding ground for fast multiplying mosquitoes, and other insects. Areas include traditionally uncovered gutters, and also terraces of high rise buildings, and shades above windows (popularly known as chhajja)-areas that are hard to reach and access. In this paper we propose the use of a quadcopter to inspect such areas and identify stagnant water patches. Water being specular in nature tends to confound traditional imageprocessing methods. Further the use of a non-traditional camera mounted on a quadcopter presents new challenges. We provide methods to get past such hurdles.
Background subtraction is an important preprocessing technique for a wide variety of problems in computervision including automatic video surveillance, anomaly detection etc. Our focus is on background subtraction of...
详细信息
ISBN:
(纸本)9781467385640
Background subtraction is an important preprocessing technique for a wide variety of problems in computervision including automatic video surveillance, anomaly detection etc. Our focus is on background subtraction of videos taken from stationary cameras. We use sparse representation and compressive sensing to propose a novel algorithm that separate the background image and present the foreground objects in each frame. Our method is robust to dynamic background scenario where the background changes with time. We also point towards the fact that our algorithm is highly parallalizable and so can subtract background in real time. We demonstrate the superiority of our method against Mixture of Gaussian, KDE model and Monnnet's method. Also our method is on par with AdaDGS in terms of visual result.
This paper presents a system for unconstrained handwritten Odia text recognition using Hidden Markov Model (HMM) framework. Existing literature for Odia text recognition works primarily with individual isolated charac...
详细信息
ISBN:
(纸本)9781467385640
This paper presents a system for unconstrained handwritten Odia text recognition using Hidden Markov Model (HMM) framework. Existing literature for Odia text recognition works primarily with individual isolated characters. In this study we introduce a Odia dataset of word samples collected from different professionals. Concavity feature from each word image is extracted in our approach. Next, the features are fed to HMM-based sequential classifier for recognition. The experiment has been performed on a large dataset consisting of 4000 words and results obtained are encouraging.
In the present work, we propose a framework for retrieving metric information of featureless cylindrical pellet imaged through stereo vision. Considering the isometry property, relative affine structure which is an in...
详细信息
ISBN:
(纸本)9781467385640
In the present work, we propose a framework for retrieving metric information of featureless cylindrical pellet imaged through stereo vision. Considering the isometry property, relative affine structure which is an invariant that depends on the depth of the point is employed for obtaining the information. This method skips considering vanishing points for depth estimation, and hence the errors are reduced remarkably as no points are assumed to be at infinity. Real world measurements are taken and the metrics of a selected reference point is assumed to be known. With these minimal information the mapping between the images using the relative affine structure is carried out that enables the 3D reconstruction. This work also blends the concepts of 360 degree rotational symmetry and orthogonal planes with the stereo vision that has minimized the error percentage. The results presented by the theory will have an impact on the design of 3D reconstruction systems for computervision and its applications.
Automatic image annotation is the computervision task of assigning a set of appropriate textual tags to a novel image. The aim is to eventually bridge the semantic gap of visual and textual representations with the h...
详细信息
ISBN:
(纸本)9781467385640
Automatic image annotation is the computervision task of assigning a set of appropriate textual tags to a novel image. The aim is to eventually bridge the semantic gap of visual and textual representations with the help of these tags. This also has applications in designing scalable image retrieval systems and providing multilingual interfaces. Though a wide varieties of powerful machine learning algorithms have been explored for the image annotation problem in the recent past, nearest neighbor techniques still yield superior results to them. A challenge ahead of the present day annotation schemes is the lack of sufficient training data. In this paper, an active Learning based image annotation model is proposed. We leverage the image-toimage and image-to-tag similarities to decide the best set of tags describing the semantics of an image. The advantages of the proposed model includes: (a). It is able to output the variable number of tags for images which improves the accuracy. (b). It is effectively able to choose the difficult samples that needs to be manually annotated and thereby reducing the human annotation efforts. Studies on Corel and IAPR TC-12 datasets validate the effectiveness of this model.
Wavelet transform of the image generates the different components basically classified in the approximation and detail components. The approximation component has major information. In this paper a partial encryption ...
详细信息
ISBN:
(纸本)9781467385640
Wavelet transform of the image generates the different components basically classified in the approximation and detail components. The approximation component has major information. In this paper a partial encryption technique is used, using only approximation component. In this paper a random array is generated which is XORed with the approximation component. The inverse transform of the matrix generated by this operation generates the encrypted image. The random array, the wavelet used and level of wavelet transform jointly serves as the key for the decryption process. In decryption is just reverse of the encryption steps.
This paper proposes a novel recommendation engine to suggest coordinated outfits to the users that complements each other. The proposed recommendation model encodes subjective knowledge of clothing experts in Multimed...
详细信息
ISBN:
(纸本)9781467385640
This paper proposes a novel recommendation engine to suggest coordinated outfits to the users that complements each other. The proposed recommendation model encodes subjective knowledge of clothing experts in Multimedia Web Ontology Language (MOWL) and makes use of evidential and causal reasoning scheme to deal with the media properties of concepts. Our approach automatically identifies the user visual personality and interprets the contextual meaning of media features of the garments in the context of input query image. As a result, personalized complementary garments based on occasion of wear are recommended to the user. We have validated our approach with garment preferences of various models with a large collection of shirts and trousers, collected from various websites.
This paper describes a sparse representation based approach to learn a classifier for assessing the video quality without a reference. First we calculate the natural scene statistics (NSS) based spatial features of ea...
详细信息
ISBN:
(纸本)9781467385640
This paper describes a sparse representation based approach to learn a classifier for assessing the video quality without a reference. First we calculate the natural scene statistics (NSS) based spatial features of each frame/ image and then learn a dictionary by K-SVD algorithm from NSS features of correct frames. In this work we identified the fact that correct frame can be represented precisely in terms of dictionary atoms but while representing a distorted frame, the error drastically increases with increase in distortion thus we can easily classify the frames as correct and distorted based on error score calculated by sparse representation framework. This framework has been validated on two datasets and we observe improved accuracies as compared to state-of-art algorithms.
暂无评论