Information Of Keypoints Research Articles

Distortion Correction using Enhanced Feature Extraction and Classification

With the advance in technology, different types of machines are used to acquire the images. These sensors acquire information and make it in the form of images. Sometime these images are affected by some distortion like barrel and pincushion. This is because theses sensors have a mark focus on either centre or edge. As it focuses on one point so it cannot be removed but can be corrected after acquiring samples. A number of methods have been used to correct this type of distortion. The previous work which we took under consideration was to collect information by extracting some texture features using that feature to classify the image which will provide the correct information at some sort of point. So, there is still need to improve the results because texture feature. But also edge feature and key point information using different algorithms. Then collect information by classifying them using neural classifier which achieves 93% accuracy. General Terms Distortion Correction, classification.

Creating Integrated Histology Flash Cards with Clinical Correlations in Teaching Histology

Well designed histology flash cards are a useful tool for learning histology. Integrating pathological images and clinical cases makes flash cards especially helpful in enhancing students’ understanding of how the normal tissue could develop into abnormal tissue, and what pathological condition and clinical disease may be associated with this change. The integrated flash cards provide students a good platform to maximize their exposure to clinical information in their early stage of medical training. Because of the portability of the flash cards, they serve as convenient learning tools for quick review, self‐assessment, and preparation for exams. We created customized histology flash cards at four levels: (1) Introductory cards that help students to quickly review basic organization of the materials; (2) Cards with normal and pathological images to help students to appreciate their differences; (3) Cards with key point information to strengthen their knowledge base, and (4) Cards with practical questions and clinical vignette in USMLE style for students to prepare for various exams. In brief, Histology Flash Cards with Clinical Correlations serve as an efficient, convenient and flexible tool for students to learn histology, as well as strengthen their ability to integrate histology and related clinical information.Grant Funding Source: University of Mississippi Medical Center

Visual content representation using semantically similar visual words

Local feature analysis of visual content, namely using Scale Invariant Feature Transform (SIFT) descriptors, have been deployed in the ‘bag-of-visual words’ model (BVW) as an effective method to represent visual content information and to enhance its classification and retrieval. The key contributions of this paper are first, a novel approach for visual words construction which takes physically spatial information, angle, and scale of keypoints into account in order to preserve semantic information of objects in visual content and to enhance the traditional bag-of-visual words, is presented. Second, a method to identify and eliminate similar key points, to form semantic visual words of high quality and to strengthen the discrimination power for visual content classification, is given. Third, an approach to discover a set of semantically similar visual words and to form visual phrases representing visual content more distinctively and leading to narrowing the semantic gap is specified.

Study on a Vision Sensing System for the Parameter Estimation of a Serial Construction Robot

In this paper, a vision sensing system was developed in order to measure the gesture of a serial construction robot and the cylinder displacement of each actuator. The system consists of a single CCD video camera and several light sources which were used to acquire the distance information of key points of the construction robot. The three-dimensional coordinates of key points were obtained from a two-dimensional video image by using image processing. The gesture of the construction robot and the cylinder displacement of each hydraulic actuator were calculated via the inverse kinematics method. In addition, the validity of the proposed method was confirmed through an experiment of parameter estimation for the friction characteristics of a hydraulic cylinder.

SIFT and shape information incorporated into fluid model for non-rigid registration of ultrasound images

Non-rigid registration of ultrasound images takes an important role in image-guided radiotherapy and surgery. Intensity-based method is popular in non-rigid registration, but it is sensitive to intensity variations and has problems with matching small structure features for the existence of speckles in ultrasound images. In this paper, we develop a new algorithm integrating the intensity and feature of ultrasound images. Both global shape information and local keypoint information extracted by scale invariant feature transform (SIFT) are incorporated into intensity similarity measure as the body force of viscous fluid model in a Bayesian framework. Experiments were performed on synthetic and clinical ultrasound images of breast and kidney. It is shown that shape and keypoint information significantly improves fluid model for non-rigid registration, especially for alignment of small structure features in accuracy.

High-level Specification and Animation of Communicative Gestures

This paper describes a complete system for the specification and the generation of visual communicative gestures. A high-level language for the specification of hand–arm communication gestures was developed. This language is based on both a discrete description of space, and a movement decomposition inspired by sign language gestures. Communication gestures are represented by symbolic commands that can be described by qualitative data, and translated in terms of spatio-temporal targets that drive the generation system. This approach is well-suited for a class of generation models controlled by key-points information at the trajectory level. The animation model used in our approach is composed of a set of sensory-motor control loops. Each of these models computes in real-time updated angular coordinates of the articulatory structure from the minimization of the distance between target and current locations. At the same time, psycho-motor laws of biological movement are satisfied. The whole control system is applied to the synthesis of communication and sign language gestures. A synthetic character is animated and some results are presented.