Summary form only given. The ESPRIT project VIEWS (Visual Inspection and Evaluation of Wide-area Scenes) (project 2152) demonstrates computer vision systems for real-time surveillance of outdoor dynamic scenes in cons...
详细信息
Summary form only given. The ESPRIT project VIEWS (Visual Inspection and Evaluation of Wide-area Scenes) (project 2152) demonstrates computer vision systems for real-time surveillance of outdoor dynamic scenes in constrained situations. Two domains, viz. ground traffic at airports and road traffic have been selected for demonstration. For applications within these domains, inspection refers to a monitoring activity which provides a description of the scene in spatio-temporal terms, while evaluation refers to the higher-level activity of surveillance which produces conceptual descriptions of the scene of meaning to the end user. Both levels of description are derived primarily from image sequence data but other, nonvisual, data may also be input. The term wide-area refers not only to the spatial extent of the possible interactions but also to their number, for example a large network of roads or a junction in an urban area, or the whole or part of the airside of an airport.< >
The authors have developed four applications where registration of images from different modalities provides opportunities for synergism. By combining NR and SPECT images of the brain, it is possible to locate functio...
详细信息
The authors have developed four applications where registration of images from different modalities provides opportunities for synergism. By combining NR and SPECT images of the brain, it is possible to locate functional abnormalities indicated by SPECT with respect to the anatomy imaged by MR. This may provide opportunities for more precise surgery of functional lesions which are not visible in MR (such as certain types of epilepsy). The combination of MR and SPECT images of the heart enables heart function to be related to heart anatomy. Combining MR and CT images of the skull base region enables surgeons to more accurately relate bony anatomy visible in CT with soft tissue features most clearly visible in MR images, and hence aids planning of complicated surgical procedures such as transtemporal resections. Combining 3D reconstructions of blood vessels with MR soft tissue images provides surgeons with accurate information about the position of blood vessels, and hence may enable mortality and morbidity caused by accidental damage to blood vessels to be reduced.< >
We are studying methods that operate in the spatial domain in which the edges can be easily identified and treated appropriately. Here we present two techniques. Each uses some form of segmentation either on the inter...
详细信息
We are studying methods that operate in the spatial domain in which the edges can be easily identified and treated appropriately. Here we present two techniques. Each uses some form of segmentation either on the interframe signal, the intraframe signal or both, combined with some form of interpolation. The two techniques will be presented, and their relative merits discussed, together with a comparison with the DCT. The comparisons show that although in high bit rate coding the transform technique performs generally better, in low bit rate applications such as videotelephony the segmentation techniques produce more acceptable results. This is because, unlike DCT coding, segmented coding preserves the perceptually important image edges, which form a major part of the structure of the image, particularly at very low bit rates.
This paper illustrates statistical techniques which can be applied to acoustic imagery to compensate for the lack of multispectral information. Fractal segmentation appears to be a very powerful technique for discrimi...
详细信息
This paper illustrates statistical techniques which can be applied to acoustic imagery to compensate for the lack of multispectral information. Fractal segmentation appears to be a very powerful technique for discriminating local variations in the sea-floor, which is important in enabling boundaries between different regions to be delineated. The measurement of spectral attenuation in the sub-bottom provides a possible tool for obtaining information on geotechnical properties of the sediments. This type of information could reduce the requirement for seabed coring surveys which are very expensive and time consuming.
In the weaving industry a variety of meshes are manufactured, using both metallic and nonmetallic materials, for a wide variety of filtration applications. At present the quality of the mesh is monitored at several st...
详细信息
In the weaving industry a variety of meshes are manufactured, using both metallic and nonmetallic materials, for a wide variety of filtration applications. At present the quality of the mesh is monitored at several stages of the manufacturing process to keep selected parameters within specified limits. This involves stopping the process approximately every half hour, manually assessing the mesh quality, and taking remedial action as necessary. There is a need to replace these manual checks with an automatic system that can continually monitor the mesh as it is being manufactured and, with appropriate actuators, control the loom on-line. This paper presents initial results of research being conducted at Sheffield University to develop such a system for wire mesh. Early experiments proved that a contact measurement technique is inappropriate due to the severe vibrations and backlash nonlinearities present in most looms. The approach adopted here therefore, is to use imageprocessingtechniques. The instrument is transputer based so that the imageprocessing can be performed on several processors to achieve the real-time performance necessary for on-line monitoring and control purposes.
Considers the development and evaluation of real-time imageprocessing algorithms capable of measuring, to sub-pixel accuracy, the displacement between images subjected to simultaneous translation and rotation. The te...
详细信息
Considers the development and evaluation of real-time imageprocessing algorithms capable of measuring, to sub-pixel accuracy, the displacement between images subjected to simultaneous translation and rotation. The techniques could have a number of applications, for example, in stabilising imagery recorded from vibrating platforms or in the data compression of moving imagery. The following classes of algorithm were investigated: phase correlation algorithms (both 2-dimensional and 1-dimensional); and intensity gradient based algorithms.< >
Presents some results of using fractals and fractal ideas in both the modelling and analysis of textures. Much of the work is directed at applications related to studies of the seabed. The real data arrives in the for...
详细信息
Presents some results of using fractals and fractal ideas in both the modelling and analysis of textures. Much of the work is directed at applications related to studies of the seabed. The real data arrives in the form of sonar or video data. The sonar data is digitised and stored and analysed in image form. The data is complex and occupies large amounts of computer storage. There is a requirement to automatically segment different areas of texture, since the manual methods are slow and error prone. The texture is related to the composition of the surface sediments, and also the change in texture with time is useful as an aid to navigation.< >
Reviews the development of motion estimation and compensation techniques in the context of efficient image coding and display and mentions typical application areas. By and large, there has been relatively little work...
详细信息
Reviews the development of motion estimation and compensation techniques in the context of efficient image coding and display and mentions typical application areas. By and large, there has been relatively little work reported on the problem of general motion, involving rotation and scale change, and the assumption of simple translation only has been widely invoked. Other areas relevant to the present discussion are those of image analysis and object tracking, where a true motion estimate is required but any kind of associated compensation operation unnecessary. One or two areas where the technique is currently employed are introduced, and to indicate selected references are indicated in the literature.< >
Motion compensation nowadays plays an important role in image sequence coding, especially in low bit-rate image sequence coding such as video telephone and high definition TV. A number of motion estimation algorithms ...
详细信息
Motion compensation nowadays plays an important role in image sequence coding, especially in low bit-rate image sequence coding such as video telephone and high definition TV. A number of motion estimation algorithms have been developed and can be roughly categorized as differential, pel-recursive, or block matching algorithms. These algorithms must, of course, be aided by some coding scheme to realize practical efficient image sequence coding, and, although transform coding is in many respects a successful image data compression technique, it has been shown to fail in further compressing motion compensated image data. The authors propose a motion compensated image sequence coding scheme with scene segmentation, post-estimation processing, and vector quantization, which reduces the bit rate of a 256*256 'Claire' sequence to 0.4 bit per pixel (bpp) with excellent, and 0.1 bpp with good, picture quality.< >
Most literature on fractals defined by self-affine systems (SASs) assume that Monte Carlo methods will be used for rendering the attractor. These are so inefficient as to be unsuitable for use in decompressing fractal...
详细信息
Most literature on fractals defined by self-affine systems (SASs) assume that Monte Carlo methods will be used for rendering the attractor. These are so inefficient as to be unsuitable for use in decompressing fractally coded images. The authors introduce a deterministic algorithm, minimum point plotting (MPP), which is optimal in the number of pixels plotted on a graphics screen without requiring an auxiliary frame store. Its efficiency in rendering simple test fractals on a sequential computer is compared to the Monte Carlo method. A generalization of the algorithm, called the Generation Game (G/sup 2/), is readily shown to be a variant of the previously described image transformation algorithm (ITA), using different initial conditions. MPP is superior to ITA for sequential machines, and G/sup 2/ is at worst equivalent for parallel applications. A further generalization of G/sup 2/ is the constant mass algorithm (CMA), which can be used to render colour and grey scale images, and comparisons of CMA with Monte Carlo methods are also given.< >
暂无评论