Genetic Learning For Adaptive Image Segmentation
Download Genetic Learning For Adaptive Image Segmentation PDF/ePub or read online books in Mobi eBooks. Click Download or Read Online button to get Genetic Learning For Adaptive Image Segmentation book now. This website allows unlimited access to, at the time of writing, more than 1.5 million titles, including hundreds of thousands of titles in various foreign languages.
Genetic Learning for Adaptive Image Segmentation
Author: Bir Bhanu
language: en
Publisher: Springer Science & Business Media
Release Date: 1994-09-30
Image segmentation is generally the first task in any automated image understanding application, such as autonomous vehicle navigation, object recognition, photointerpretation, etc. All subsequent tasks, such as feature extraction, object detection, and object recognition, rely heavily on the quality of segmentation. One of the fundamental weaknesses of current image segmentation algorithms is their inability to adapt the segmentation process as real-world changes are reflected in the image. Only after numerous modifications to an algorithm's control parameters can any current image segmentation technique be used to handle the diversity of images encountered in real-world applications. Genetic Learning for Adaptive Image Segmentation presents the first closed-loop image segmentation system that incorporates genetic and other algorithms to adapt the segmentation process to changes in image characteristics caused by variable environmental conditions, such as time of day, time of year, weather, etc. Image segmentation performance is evaluated using multiple measures of segmentation quality. These quality measures include global characteristics of the entire image as well as local features of individual object regions in the image. This adaptive image segmentation system provides continuous adaptation to normal environmental variations, exhibits learning capabilities, and provides robust performance when interacting with a dynamic environment. This research is directed towards adapting the performance of a well known existing segmentation algorithm (Phoenix) across a wide variety of environmental conditions which cause changes in the image characteristics. The book presents a large number of experimental results and compares performance with standard techniques used in computer vision for both consistency and quality of segmentation results. These results demonstrate, (a) the ability to adapt the segmentation performance in both indoor and outdoor color imagery, and (b) that learning from experience can be used to improve the segmentation performance over time.
Soft Computing for Image Processing
Any task that involves decision-making can benefit from soft computing techniques which allow premature decisions to be deferred. The processing and analysis of images is no exception to this rule. In the classical image analysis paradigm, the first step is nearly always some sort of segmentation process in which the image is divided into (hopefully, meaningful) parts. It was pointed out nearly 30 years ago by Prewitt (1] that the decisions involved in image segmentation could be postponed by regarding the image parts as fuzzy, rather than crisp, subsets of the image. It was also realized very early that many basic properties of and operations on image subsets could be extended to fuzzy subsets; for example, the classic paper on fuzzy sets by Zadeh [2] discussed the "set algebra" of fuzzy sets (using sup for union and inf for intersection), and extended the defmition of convexity to fuzzy sets. These and similar ideas allowed many of the methods of image analysis to be generalized to fuzzy image parts. For are cent review on geometric description of fuzzy sets see, e. g. , [3]. Fuzzy methods are also valuable in image processing and coding, where learning processes can be important in choosing the parameters of filters, quantizers, etc.
Tetrobot
Author: Gregory J. Hamlin
language: en
Publisher: Springer Science & Business Media
Release Date: 2013-03-09
Robotic systems are characterized by the intersection of computer intelligence with the physical world. This blend of physical reasoning and computational intelligence is well illustrated by the Tetrobot study described in this book. Tetrobot: A Modular Approach to Reconfigurable Parallel Robotics describes a new approach to the design of robotic systems. The Tetrobot approach utilizes modular components which may be reconfigured into many different mechanisms which are suited to different applications. The Tetrobot system includes two unique contributions: a new mechanism (a multilink spherical joint design), and a new control architecture based on propagation of kinematic solutions through the structure. The resulting Tetrobot system consists of fundamental components which may be mechanically reassembled into any modular configuration, and the control architecture will provide position control of the resulting structure. A prototype Tetrobot system has been built and evaluated experimentally. Tetrobot arms, platforms, and walking machines have been built and controlled in a variety of motion and loading conditions. The Tetrobot system has applications in a variety of domains where reconfiguration, flexibility, load capacity, and failure recovery are important aspects of the task. A number of key research directions have been opened by the Tetrobot research activities. Continuing topics of interest include: development of a more distributed implementation of the computer control architecture, analysis of the dynamics of the Tetrobot system motion for improved control of high-speed motions, integration of sensor systems to control the motion and shape of the high-dimensionality systems, and exploration of self-reconfiguration of the system. Tetrobot: A Modular Approach to Reconfigurable Parallel Robotics will be of interest to research workers, specialists and professionals in the areas of robotics, mechanical systems and computer engineering.