The introduction of intelligence near each photosensitive element in focal plane arrays (FPA) leads to sensory devices-called artificial retinas-which may no longer output raw images but, rather, much more concentrate...
详细信息
The introduction of intelligence near each photosensitive element in focal plane arrays (FPA) leads to sensory devices-called artificial retinas-which may no longer output raw images but, rather, much more concentrated forms of information. In particular, when the on-sensor imageprocessing facilities are powerful enough to allow some structural patternrecognition, lists of pixels of interest become an output format of choice from retina to microprocessor. This implies the development of specific output techniques and operators to be integrated in the focal plane. After an in-depth presentation of the motivations in the context of programmable artificial retinas (PAR) for robot vision, two original solutions to the problem are presented, corresponding to two different trade-offs between efficiency and VLSI implementation cost. The first one is a compact hardware solution, which allows to sense pixels of interest from the sides of the 2D pixel array. The second one, a mainly software technique, exploits the mathematical concept of de Bruijn arrays for a distributed encoding of pixel addresses on the PAR.
We present a perception system, called Artisan, that semi-automatically builds 3-D models of a robot's workspace. Range images are acquired with a scanning laser rangefinder and then processed, based an a systemat...
详细信息
We present a perception system, called Artisan, that semi-automatically builds 3-D models of a robot's workspace. Range images are acquired with a scanning laser rangefinder and then processed, based an a systematic sensor characterization, to remove noise and artifacts. Complex 3-D objects represented as surface meshes are subsequently recognized in the range images and inserted into a virtual workspace. This graphical virtual workspace is then used to by human operators to plan and execute remote robotic operations.
We address in this paper the problem of defining fuzzy relationships between objects for patternrecognition purposes. We distinguish two kinds of spatial relationships. Some of them are well defined if the objects ar...
详细信息
ISBN:
(纸本)0819423599
We address in this paper the problem of defining fuzzy relationships between objects for patternrecognition purposes. We distinguish two kinds of spatial relationships. Some of them are well defined if the objects are crisp, like adjacency, inclusion or other set relationships. But since they are highly sensitive to errors or imprecision in segmentation, more useful measures can be obtained by fuzzifying these concepts. We define such measures using fuzzification principles or direct translation of binary equations into fuzzy ones. Other relationships are inherently vague concepts, like relative position. Fuzzy definitions of such relationships are then more consistent than crisp ones. We propose definitions of relative position between crisp or fuzzy objects, and illustrate them on buildings in aerial images.
Proceedings (12 reports) of the conference on Non-Conventional pattern Analysis in remotesensing are presented. The main topics discussed at the conference were following: neural networks for geographic information p...
详细信息
Proceedings (12 reports) of the conference on Non-Conventional pattern Analysis in remotesensing are presented. The main topics discussed at the conference were following: neural networks for geographic information processing;algorithms for supervised classification of remotesensingimages;fuzzy logic and neural techniques integration;numeric and symbolic data fusion as an approach to remotesensingimage analysis;incorporating mixed pixels in supervised classification development and an approach to fuzzy land cover mapping.
In the framework of an image interpretation system for automatic cartography based on remotesensingimage classification improved by a photo interpreter knowledge, we developed a system based on neural networks which...
详细信息
ISBN:
(纸本)0819423599
In the framework of an image interpretation system for automatic cartography based on remotesensingimage classification improved by a photo interpreter knowledge, we developed a system based on neural networks which simultaneously produce fuzzy rules, with their linguistic approximation as well as final classification. This paper describes the succession of steps used with this aim in view. Particularly it investigates the application of mutual information criteria to simplify fuzzy rules.
This paper presents an automatic method for processing digitized images of cadastral maps. The method includes two major algorithms: a segmentation and a Raster-to-Vector conversion. Those algorithms use a simple data...
详细信息
This paper presents an automatic method for processing digitized images of cadastral maps. The method includes two major algorithms: a segmentation and a Raster-to-Vector conversion. Those algorithms use a simple data-list structure for recording data created during single-pass, row-majority scanning and line tracing. The segmentation algorithm obtains the positions and sizes of symbols and characters, in addition to completing map segmentation and proving useful for patternrecognition. The Raster-to-Vector conversion algorithm obtains topological information necessary to relate cadastral map spatial data to line start points, midpoints, intersection points, and termination points. It consists of four integrated sub-algorithms that remove noise, unify run-length coordinates, and perform synchronous line approximations and logical linkage of line breaks. Straight, angled, and curved lines can then be completely reconstructed for display. Also presented are six indices that verify algorithm and experimental results.
imageprocessing applications and especially those in the area of remotesensing are often characterized by a high degree of complexity. We introduce a general framework, called 'active fusion', that actively ...
详细信息
A pixel is not an atom. Most imageprocessing algorithms basically assume that a pixel is not dividable;in other words, its inside is homogeneous consisting of only one classification class. This assumption is, howeve...
详细信息
The scattering process of electromagnetic waves is dominated by the match between wavelength and the geometric dimensions of surface structures. With respect to the microwave radar bands millimeter-waves are better ma...
详细信息
ISBN:
(纸本)0819421235
The scattering process of electromagnetic waves is dominated by the match between wavelength and the geometric dimensions of surface structures. With respect to the microwave radar bands millimeter-waves are better matched to small surface features of terrain. Therefore this frequency band is able to gain additional information on the terrain of interest. For high resolution imaging SAR is the favorite solution also for millimeter-wave frequencies. Compared to more classical radar bands millimeter-waves offer advantages in the SAR processing, because due to the higher primary resolution at a given antenna aperture sources of image distortions such as range migration or depth of focus can be neglected at these frequencies. Moreover the inherently short aperture time for a given resolution improves the relation to the time constant of flight instabilities and makes motion compensation a simple process. A coherent, polarimetric, high range resolution radar, operating at a nominal frequency of 94 GHz, has been installed onboard an aircraft to allow remotesensing measurements in a side looking synthetic aperture approach. The radar-raw-data were registered together with time code and inertial data of the aircraft and later on evaluated by an off-line SAR-processor. The resulting images then had to undergo an automatic recognition process to extract certain complex targets using a knowledge based production system. The paper describes the measurement system and discusses the evaluation procedures with emphasis on the applied SAR algorithm. Examples of radar images at 94 GHz are shown and samples of patternrecognition derived from the SAR images are shown.
The interpretation of airphotos relies heavily on the identification of textures and spatial patterns. The 2-D wavelet transform can be used to quantify simple patterns for automated classification of pixels in an ima...
详细信息
ISBN:
(纸本)0819422061
The interpretation of airphotos relies heavily on the identification of textures and spatial patterns. The 2-D wavelet transform can be used to quantify simple patterns for automated classification of pixels in an image. The transform generates a set of images similar in format to a multispectral image deck, but based on spatial information localized about each pixel. The images in this ''wavelet'' deck each correspond to an analysis of patterns in the original image at different spatial resolutions. For example, where one image in the wavelet deck depends on spatial variations within a 32 x 32 pixel area the next image in the deck contains information on spatial features within a 16 x 16 pixel area Individual pixels in the wavelet deck can be classified using the same classification and patternrecognition algorithms used with multispectral images. Classifications based on the wavelet deck of four airphoto samples using a minimum-distance to mean algorithm and an artificial neural network are presented.
暂无评论