311 research outputs found
Novel Image Compression and Deblocking Approach Using BPN and Deep Neural Network Architecture
No full textPart 6: Vision CognitionInternational audienceMedical imaging is an important source of digital information to diagnose the illness of a patient. The digital information generated consists of different modalities that occupy more disk space, and the distribution of the data occupies more bandwidth. A digital image compression technique that can reduce an image's size without losing much of its important information is challenging. In this paper, a novel image compression technique based on BPN and Arithmetic coders is proposed. The high non-linearity and unpredictiveness of the interrelationship between the pixels present in the image to be compressed is handled by BPN. An efficient coding technique called Arithmetic coding is used to produce an image with a better compression ratio and lower redundancy. A deep CNN based image deblocker is used as a post-processing step to remove the artefacts present in the reconstructed image to improve the quality of the reconstructed image. The effectiveness of the proposed methodology is validated in terms of PSNR. The proposed method is able to achieve about a 3% improvement in PSNR compared with the existing methods
Comparative Analysis of Machine Learning Algorithms for Categorizing Eye Diseases
No full textPart 9: Medical Artificial IntelligenceInternational audienceThis paper presents a comparative study on different machine learning algorithms to classify retinal fundus images of glaucoma, diabetic retinopathy, and healthy eyes. This study will aid the researchers to know about the reflections of different algorithms on retinal images. We attempted to perform binary classification and multi-class classification on the images acquired from various public repositories. The quality of the input images is enhanced by using contrast stretching and histogram equalization. From the enhanced images, features extraction and selection are carried out using SURF descriptor and k-means clustering, respectively. The extracted features are fed into perceptron, linear discriminant analysis (LDA), and support vector machines (SVM) for classification. A pretrained deep learning model, AlexNet is also used to classify the retinal fundus images. Among these models, SVM is trained with three different kernel functions and it does multi-class classification when it is modelled with Error Correcting Output Codes (ECOC). Comparative analysis shows that multi-class classification with ECOC-SVM has achieved high accuracy of 92%
A Framework for the Approximation of Relations
No full textPart 2: Uncertain TheoryInternational audienceThe paper proposes a foundation to the approximation of relations by means of relations. We discuss necessary, possible and sufficient approximations and show their links with other topics, such as refinement and simulation. The operators introduced in the paper has been tested on computers
Some Discussions on Subjectivity of Machine and its Function
No full textPart 1: Brain CognitionInternational audienceMany people think that, in order to improve the capabilities of a machine, the machine should have its initiative. Consequently, in order to make machine having its initiative, the consciousness of machine should be established. Here, we propose that instead of consciousness, the subjectivity of the machine should be considered. Thus we can avoid unnecessary arguments, and conduct fruitful discussions. We analyze the subjectivity of a machine and propose a working definition, and discuss 4 major aspects of subjectivity, namely, a priori and cogitating, active perception to outside, self awareness, and dynamic action. Dynamic action of a machine is crucially important and we suggest a possible way to approach it
Supplier selection in Telecom supply chain management: a Fuzzy-Rasch based COPRAS-G method
No full textIn the past decade, global competition are forcing firms to increase their level of outsourcing for raw or semi-finished products and building long term relationship with their supply chain partners. The objective is to present a wide-ranging decision making technique for ranking supplier alternatives in view of the effect of selected criteria. A proposed method is developed aiming the usage of Fuzzy-Rasch model applying five point Likert scale for criteria weight and Grey based COmplex PRoportional ASsessment (COPRAS-G) method for evaluating and ranking the potential alternatives, as per criteria. The applicability of the induced methodology for supplier selection problem in all environments is shown through a case study in telecommunication sector. A sensitivity analysis is performed based on changing weight patterns of criteria to show the stability in ranking result of the proposed approach. Further, a comparative analysis between the ranking results of proposed method done with existing grey multi-attribute decision-making methods viz. VIKOR-G, ARAS-G and TOPSIS-G using spearman’s correlation coefficient for checking the reliability of the ranking result
A multi-criteria decision making for renewable energy selection using Z-numbers in uncertain environment
No full textIn recent era of globalization, the world is perceiving an alarming rise in its energy consumption resulting in shortage of fossil fuels in near future. Developing countries like India, with fast growing population and economy, is planning to explore among its existing renewable energy sources to meet the acute shortage of overall domestic energy supply. For balancing diverse ecological, social, technical and economic features, selection among alternative renewable energy must be addressed in a multi-criteria context considering both subjective and objective criteria weights. In the proposed COPRAS-Z methodology, Z-number model fuzzy numbers with reliability degree to represents imprecise judgment of decision makers’ in evaluating the weights of criteria and selection of renewable energy alternatives. The fuzzy numbers are defuzzified and renewable energy alternatives are prioritized as per COmplex PropoRtional ASsessment (COPRAS) decision making method in terms of significance and utility degree. A sensitivity analysis is done to observe the variation in ranking of the criteria, by altering the coefficient of both subjective and objective weight. Also, the proposed methodology is compared with existing multi-criteria decision making (MCDM) methods for checking validity of the obtained ranking result
A hybrid MCDM approach for selection of financial institution in supply chain risk management
No full textAn Induced Fuzzy Rasch-Vikor Model for Warehouse Location Evaluation under Risky Supply Chain
No full textCoevolution of cooperation and network structure in social dilemmas in evolutionary dynamic complex network
No full text- …
