The Watershed algorithm has been studied extensively, and has been applied to image segmentation due to its accuracy and robustness. However, the watershed requires a large amount of memory, and is computationally intractable for segmenting large images. In this paper, we introduce a novel hierarchical region-of-interest (ROI) detection scheme, which is used as a prelude to segmentation. With the help of the detection algorithm, watershed segmentation can be applied to the small detected regions, rather than to the entire image. Therefore, it can process large images by selectively segmenting ROIs. We focus on our new ROI detection algorithm, and how it is integrated into a system for large-image segmentation. We demonstrate the efficiency of the proposed scheme by processing a variety of images.
Temporal segmentation of video in the compressed domain is becoming increasingly popular due to its computational advantages over video decompression followed by pixel-domain segmentation. This paper discusses the advantages of compressed-domain processing, and proposes a computationally-efficient method of detecting scene changes without reconstructing the video. The target application provides requirements that allow the algorithm to avoid complicated processing that searches for unnatural scenes changes such as dissolves, fades, and wipes that are common studio effects. The paper provides experimental results to demonstrate operation of the algorithm on real data.
The development of anthropomorphic robots for simple autonomous operations such as grasping of tools requires algorithms for automatic recognition and pose estimation of the objects. Correlation-based pattern recognition offers a robust set of tools for pose-specific detection and identification of objects. This paper discusses a system-level approach to image understanding whereby a robot is provided with training data (in the form of a computer model and an associated matched filter set), is presented with a view of the target object, and is expected to indicate recognition and to calculate a six degree-of-freedom pose estimate for the object. The pose information would then be used to specify a grasping orientation for the robot's hand. Examples are given of a proof-of-concept demonstration of an approach for an anthropomorphic robot developed at the Johnson Space Center.
Authors made theoretical research in building of correlation filters, using direct decomposition of integral transforms on correspondent orthonormal basis. The obtained filters have different properties, which are quite good for different image recognition problems: 1). Such filters itself have all properties of elementary function; 2). From the family of such filters we can easily make MACE filters (authors made it for the case of wavelet transform); 3). Invariant properties of filters of this family are determined by the presence of such properties in integral transform, which is the base for construction of such filters (so we can select these properties by the selection of the type of transform). Also authors made experimental check-up of modeling of results on a computer. Results of modeling shows good perspective of such filters in solving of such problems.
In this paper, a new VHLE(volume holographic lenticular element)-based multiview autostereoscopic 3D display system is proposed and implemented. In the proposed system, the lenticular sheet which has been used in the conventional autostereoscopic 3D display system is optically implemented by using the multiplexed volume hologram, which is called a VHLE. Thus, like the conventional lenticular sheet, the VHLE can be used as an optical directional modulator in the multiview stereoscophic display system. This VHLE can be made by using the volume holographic technique in which both of angular multiplexing and space-division multiplexing are used to record the diffraction gratings in the holographic recording materials. Once the diffraction gratings are recorded in the volume hologram, then this holographic device can play the same role of the conventional lenticular sheet. In addition, a design tool for synthesizing the VHLE for the implementation of an arbitrary multiview 3D display system is also discussed. Through some experimental results, a possibility of implementing the VHLE-based multiview autostereoscopic display system is presented.
Imaging spectrometers allowing spatially resolved targets to be spectrally discriminated are valuable for remote sensing and defense applications. The drawback of such instruments is the need to quickly process very large amounts of data. In this paper we demonstrate two imaging systems which detect a dim target in a bright background, using the coherence contrast between them, generating much less data but only operating over a limited optical bandwidth. Both systems use a passband filter, a Michelson interferometer, coupling optics and a CCD camera. The first uses the interferometer in a spatial mode, by tilting one of the mirrors to create a set of line fringes on the CCD array. The visibility of these fringes is proportional to the degree of coherence. The interferogram is displayed spatially on the CCD array, as a function of the path differences. The second system uses the interferometer in a temporal mode. A coherent point target and an extended background are imaged through the interferometer onto the CCD array, and one of the interferometer's mirrors is scanned longitudinally to vary the path difference in time. In both cases the coherent target is detected over a large dynamic range down to negative signal-to-background power ratios (in dB). The paper describes an averaging technique to improve the signal-to-noise ratio and correction techniques required to extract interferograms from the images. The spatial technique developed has the advantage of using no moving parts.
In this paper, we present a region-based scalable coding technique that can be used in interactive image/video communications. This method has a capability of near lossless coding for a specific region of interest (ROI), while the rest of the region is coded with high quality lossy codec. There are many potential applications of the region-based scalable video coding method in the areas of interactive communications such as storm tracking, wild fire and air traffic monitoring using satellite still images and near real time video sequences. It enables the server/client system to reduce data traffic across the networks while the quality of a specific area of interest chosen for each client's needs is still satisfied. We tested this technique by applying it to still images and traffic monitoring image sequences. The results consistently show high level of performance.
Current research on artificial vision and pattern recognition tends to concentrate either on numerical processing (filtering, morphological, spectral) or in symbolic or subsymbolic processing (neural networks, fuzzy logic, knowledge-based systems). In this work we combine both kinds of processing in a hybrid image processing architecture. The numerical processing part implements the most usual facilities (equalization, convolution filters, morphological filters, segmentation and description) in a way adequate to transform the input image into a polygonal outline. Then recognition is performed with a rule-based system implemented in Prolog. This allows a neat high-level representation of the patterns to recognize as a set of logical relations (predicates), and also the recognition procedure is represented as a set of logical rules. To integrate the numerical and logical components of our system, we embedded a Prolog interpreter as a software component within a visual programming language. Thus, our architecture features both the speed and versatility of a visual language application, and the abstraction level and modularity of a logical description.
Consider an hypothetical image processing system, where a given target is to be identified. The usual sequence of steps consists on an image equalization to adapt to the illumination situation. Then the image is binarized, allowing a morphological filter to correct the noisy edges and shapes by means of an indeterminate sequence of openings or closings. The resulting image can then be segmented and recognized. If the results are unsatisfactory, then the processing parameters in any of the previous steps must be changed, perhaps by trial and error. For instance, the binarization threshold can be raised or lowered, and the following steps must be performed again to see the results. This is obviously cumbersome, tedious and error prone. The Image Processing Spreadsheet PDICalc is a simple but powerful combination of two different and widespread software technologies. It's benefit comes from enabling users to build an image processing pipeline, considering each step separately, and visualizing the results of modifying the parameters of each step in the final image. A spreadsheet based user interface eliminates the tedious and repetitive interaction that characterizes current image processing software. Users can build a processing template and reliably repeat often needed processing without the effort of redevelopment or recoding. In the cited example the user simply creates the processing template, defining each cell of the spreadsheet as the result of applying a given processing step on another cell. This template can be then reused with any input image, can be stored for future processing sessions, and every step can be trimmed precisely to achieve the desired results. Our implementation considers most of the image processing techniques as its building blocks. Arithmetic operators are overloaded to represent per pixel operations. We included also equalization and histogram correction, arbitrary convolution filtering, arbitrary morphological filtering (with programmed repetition), Fourier operations, and several segmentation techniques.
Microarray technology is increasingly used as a means of high throughput analysis of human, non-human and plant genomes. Manual methods of array production have inherent imperfections and variations of the quality of output data derived from these arrays. For paired microarray images acquired using manual methods of array production, image registration is necessary for aligning corresponding spots for comparison. In this research, a dynamic programming technique is investigated for registering and comparing microarray image pairs. The output of the cost function developed provides a similarity measure between the images and can be applied to evaluating the quality of the image pair. The backward solution from the dynamic programming algorithm developed provides the basis for registering the image pairs. The image registration technique provides for an optimal alignment of the image pairs. The aligned images facilitate spot-to-spot comparisons between the image pairs for detecting specific genetic expressions that could be related to bio-medical functions.
In this paper, a new stereo tracking system using the hierarchical opto-digital algorithms is proposed and implemented. In the proposed method, a moving target object is initially extracted from the input image by removing the background noises by applying the region-based SAD algorithm to the sequential left image. And then, the location coordinates of the moving target for each of the sequential input frames are extracted through performing the optical BPEJTC between the reference image of the extracted target object and the stereo input image. These extracted coordinate values are finally used for controlling the convergence angle and the pan/tilt embedded to the conventional stereo target tracking system. From some experimental results with the 20 frames of the stereo input image pairs, the proposed system is found to be able to effectively extract the area where the target object is located from the stereo input image regardless of the background noises. With the location values of the tracking object obtained from the execution of the optical BPEJTC, the convergence angle and the pan/tilt of the stereo cameras are found to be successfully controlled. Therefore, in this paper, a feasibility test for implementing the stereo tele-working system or the stereo robot vision system using the proposed algorithm is suggested.
Secondary electron microscope images (SEIs) are analyzed in an automatic system to provide material properties for fracture analyses etc. The image processing presently used is summarized and initial results presented. Our new grain boundary (GB) detection and line following algorithms are presented. Line following locates triple junctions (TJs) (points where 3 grains meet). Initial versions of advanced algorithms to calculate TJ angles, reduce false GBs and TJs and select a reduced number of good probe points are advanced using a new noise map. Registration and formation of mosaic images are also addressed. Results on about 2000 sector images in 5 slices are presented.