Search

In today's time, the research mainly focuses on designing a Chat-bot which responds to a user's query in the most efficient manner. However, state-of-the-art works on chat-bot design are unable to emotionally connect with the user and follow-up to a conversation. In this paper, a socially and emotionally active intelligent assistant, AMIBO is proposed. It detects and recognizes the face, perceives emotion via speech and vision capabilities, provides empathetic and intelligent responses. In order to enhance the experience of AMIBO, navigation and information delivery system have been integrated in the bot. The analysis of the feature vector (on 450 subjects) is done for all the initially taken 75 distances and 15 angles reduced to 26 distances and 11 angles feature vector. An encouraging accuracy of 99% was achieved on CK+ dataset and 97% on KDEF dataset. [ABSTRACT FROM AUTHOR]

The primary aim of this paper is to develop an accurate stereo matching algorithm based on cost aggregation with adaptive support weight (CAASW). In this study, we use a pair of images (from left and right cameras) to find corresponding points. First, the truncated absolute difference is represented as cost computing, and the cost aggregation is completed with adaptive support weight. The winner take all method is then used to find the minimum cost aggregation value of the location in order to obtain the initial disparity. In order to enhance the accuracy of this study, a disparity map is employed, which uses continuity for disparity neighboring relationships; the histogram is represented as a disparity refinement, making it possible to reduce the disparity map's errors. In this paper, the CAASW can be divided into two parts. The first part is CABSW, a method employing binary target and reference images with an area of intersection to form an irregular adaptive support window. The second part is CAASW, using similarity and proximity as features of an adaptive support window with CABSW. In order to better represent the accuracy of this method, the experiment uses the Middlebury database, in addition to other methods, for comparison and analysis, to explore the experimental results and to obtain results with a lower percentage of unsatisfactory matching pixels. Future research will explore applications of this method in robot navigation, industrial manufacturing, human interface, three-dimensional reconstruction and improved computer intelligence capabilities. [ABSTRACT FROM AUTHOR]

The economic, social, cultural growth of most developing countries depends upon the agriculture sector. However, plant disease may lead to inferior crop yield economic growth. Various computer vision and artificial intelligence-based schemes have been presented in past for automatic plant leaf disease detection but their performance is inadequate due to underprivileged feature representation, lower-order correlation of raw features, data imbalance problem, less generalization. This paper presents automatic plant leaf disease detection using Deep Convolutional Neural Network (DCNN) to increase the feature representation and correlation and Generative Adversarial Network (GAN) for data augmentation to cope up with data imbalance problem. Extensive experimentations are performed on ten classes of tomato plant disease from Plant Village leaf disease database. Effectiveness of proposed scheme is evaluated based on accuracy, precision, recall, F1 score which has shown momentous improvement over customary methods for plant leaf disease database (Accuracy- 99.74%, precision- 0.99, recall-0.99, F1 score -0.99). [ABSTRACT FROM AUTHOR]

The Self-Quotient Image (SQI) Method [Wang H, Li SZ, Wang Y, et al. Self quotient image for face recognition. International Conference on Image Processing (ICIP’04); 2004;Vol. 2. p. 1397-1400; Wang H, Li SZ, Wang Y. Generalized quotient image. IEEE CVPR; 2004; Vol. 2. p. 498-505] is a simple method for lighting normalization based on the Quotient Image method [Shashua A, Riklin-Raviv T. The quotient image: class-based re-rendering and recognition with varying illuminations. T Pattern Anal Mach Intel. 2001;23(2):129-139; Riklin-Raviv T, Shashua A. The quotient image: class based recognition and synthesis under varying illumination. Proceedings of the 1999 Conference on Computer Vision and Pattern Recognition; 1999; Fort Collins (CO). p. 566-571]. The main advantage of the SQI is the use of only one image for lighting normalization. Nevertheless, the SQI still has few disadvantages which make hard to use it in some face recognition systems. In this paper, we introduce the modified version of the SQI method based on globally modified Gaussian filter kernel. In this modification, we tried to solve the disadvantages of the original SQI method, simplify the computational process, and increase the quality of illumination normalization. We have investigated two modification of the original SQI method and shown how they normalize different shadow regions. [ABSTRACT FROM AUTHOR]

In recent years, human action recognition has become a key issue in all intelligence visual surveillance systems, and it is also an important area of research in computer vision. In this paper, a novel approach is presented for human action recognition. Associated sparse geometric features are extracted based on the second generation Bandelet transformation. Aiming at the latest development of Bandelet transformation, the statistical features were proposed to represent each frame of a video and information from geometric flow to characterise the image context, especially for human body shapes and inner information. Features are selected with AdaBoost algorithm, and test experiments are carried on AdaBoost and Support Vector Machines (SVM). This method is easier to be implemented in the real world, without the complex probabilistic modelling of motion patterns and no need background subtraction. Experimental results demonstrate that the proposed approach can achieve satisfying performance on the Weizmann action dataset and KTH dataset, and exhibits considerable robustness to the body's size change, because of its inherent strictness in describing human action with geometric flows. [ABSTRACT FROM AUTHOR]

One of the core challenges in developing a computer system for machine learning is to make the system learn efficiently and effectively like a real human by grasping the domain knowledge exemplified by human experts. In this challenge, we have introduced a hybrid image synthesis model that can simulate one of the human’s learning capabilities in the vision field - the ability to synthesize images of convex objects by identifying solid geometries and textures of specific objects using few photographs. We have incorporated an ontology-based, domain knowledge on solid geometries into our model to synthesize large number of training images with only a minimum number of input images. Our initial experiments have shown that our model has convincing improvements by demonstrating a substantially better FAR/FRR/EER results when it is compared with a smaller set of non-synthetic images. [ABSTRACT FROM AUTHOR]