The solution of the segmentation problem requires a mechanism for partitioning the image array into low-level entities based on a model of the underlying image structure. A piecewise-smooth surface model for image data that possesses surface coherence properties is used to develop an algorithm that simultaneously segments a large class of images into regions of arbitrary shape and approximates image data with bivariate functions so that it is possible to compute a complete, noiseless image reconstruction based on the extracted functions and regions. Surface curvature sign labeling provides an initial coarse image segmentation, which is refined by an iterative region-growing method based on variable-order surface fitting. Experimental results show the algorithm's performance on six range images and three intensity images. >

/pdf/segmentation-through-variable-order-surface-fitting-29h4kwrbaf.pdf

Segmentation through variable-order surface fitting

A general-purpose computer vision system must be capable of recognizing three-dimensional (3-D) objects. This paper proposes a precise definition of the 3-D object recognition problem, discusses basic concepts associated with this problem, and reviews the relevant literature. Because range images (or depth maps) are often used as sensor input instead of intensity images, techniques for obtaining, processing, and characterizing range data are also surveyed.

Three-dimensional object recognition

The problem of recognizing and locating rigid objects in 3-D space is important for applications of robotics and naviga tion. We analyze the task requirements in terms of what information needs to be represented, how to represent it, what kind of paradigms can be used to process it, and how to implement the paradigms. We describe shape surfaces by curves and patches, which we represent by linear primitives, such as points, lines, and planes. Next we describe algo rithms to construct this representation from range data. We then propose the paradigm of recognizing objects while locat ing them. We analyze the basic constraint of rigidity that can be exploited, which we implement as a prediction and verifi cation scheme that makes efficient use of the representation. Results are presented for data obtained from a laser range finder, but both the shape representation and the matching algorithm are general and can be used for other types of data, such as ultrasound, stereo, and tactile.

The representation, recognition, and locating of 3-d objects

There are some difficulties in applying 3-dimensional computer vision to industry. One of them is that a vast amount of computation is required for low level processing. Some special hardware systems are described and one device is shown in more detail. Applications of the hardware are discussed in three examples. Two methods for range data acquisition employing special processors are described. Then our studies for range data processing are introduced. Finally, a complete production system is proposed which makes use of unified geometric models to be shared by CAD, CAM, and visual processing. This approach can solve a second difficulty in applying 3-dimensional computer vision to industry, namely the problem of the extensive programming effort that is required.

Three-Dimensional Computer Vision

This paper presents a comparative study and survey of model-based object-recognition algorithms for robot vision. The goal of these algorithms is to recognize the identity, position, and orientation of randomly oriented industrial parts. In one form this is commonly referred to as the "bin-picking" problem, in which the parts to be recognized are presented in a jumbled bin. The paper is organized according to 2-D, 2½-D, and 3-D object representations, which are used as the basis for the recognition algorithms. Three central issues common to each category, namely, feature extraction, modeling, and matching, are examined in detail. An evaluation and comparison of existing industrial part-recognition systems and algorithms is given, providing insights for progress toward future robot vision systems.

Model-based recognition in robot vision

This paper describes an approach to the recognition of stacked objects with planar and curved surfaces. The range data of a scene are obtained by a range finder. The system works In two phases. In a learning phase, a scene containing a single object Is described In terms of properties of regions and relations between them. This description Is stored as an object model. In a recognition phase, an unknown scene Is described In the same way as In the learning phase. And then the description is matched to the object models so that stacked objects are recognized one by one. Efficient matching is achieved by a combination of data-driven and model-driven search process. Experimental results for blocks and machine parts are shown.

Object recognition using three-dimensional information

This paper describes an approach to the recognition of stacked objects with planar and curved surfaces. The system works in two phases. In the learning phase, a scene containing a single object is shown one at a time. The range data of a scene are obtained by a range finder. The description of each scene is built in terms of properties of regions and relations between them. This description is stored as an object model. In the recognition phase, an unknown scene is described in the same way as in the learning phase. Then the description is matched to the object models so that the stacked objects are recognized sequentially. Efficient matching is achieved by a combination of data-driven and model-driven search processes. Experimental results for blocks and machine parts are shown.

Object Recognition Using Three-Dimensional Information

A scene description method using three-dimensional information

As one of major tankers routes in Indonesia, Malacca Strait is potentially prone to oil spill pollution. SAR image data is used to detect oil slick on sea surface, because of its capability for large scale of sea monitoring, and to solve clouds covering problem in Indonesia. Oil slick is visible on SAR images as dark patches, because it decreases the radar backscatter on sea surface, which is explained by Bragg scattering theory. This project proposes the procedure for auto segmentation of oil slick in SAR image data. To reduce speckle in SAR image, we used the Bayesian approach with maximum a posteori filter with assumption that radar reflectivity and speckle noise follow Gamma distribution. This technique is compared with another adaptive filter such as Lee and Frost filter, which shows the best result on reducing speckle on whole area with the lowest ratio of mean and standard deviation, reducing the bright fleck on oil area, and showing the continuity on oil slick and sea area, which is make easier in feature extraction of oil slick. Maximum entropy technique is used for feature extraction on oil slick segmentation with assumption that only two moments are to be determined i.e. oil slick and sea area. It shows the best performance on oil slick segmentation among the others co-occurrence techniques. The area classification between oil slick and look-alike area on low wind area is done based on these result. Usually, low wind areas are located near coastal line. Therefore, coastal line is applied as boundary condition. Coastal line is detected with Canny filter using first derivative of Gaussian as edge detector and mark the position of edge where the gradient is local maximum. Finally, if areas are captured near coastal lines, these areas are marked as the possible looks-alike area.

Masaki Oshima

Papers

Object recognition using three-dimensional information

Object Recognition Using Three-Dimensional Information

A scene description method using three-dimensional information

Auto Segmentation of Oil Slick in RADARSAT SAR Image Data around Rupat Island, Malacca Strait

Multi Light Sources Range Finder System.