An artificial neural network that self-organizes to recognize various images presented as a training set is described. One application of the network uses multiple functionally disjoint stages to provide pattern recognition that is invariant to translations of the object in the image plane. The general form of the network uses three stages that perform the functionally disjoint tasks of preprocessing, invariance, and recognition. The preprocessing stage is a single layer of processing elements that performs dynamic thresholding and intensity scaling. The invariance stage is a multilayered connectionist implementation of a modified Walsh-Hadamard transform used for generating an invariant representation of the image. The recognition stage is a multilayered self-organizing neural network that learns to recognize the representation of the input image generated by the invariance stage. The network can successfully self-organize to recognize objects without regard to the location of the object in the image field and has some resistance to noise and distortions.>

A method that maximizes the information flow through a constrained communications channel when it is desired to estimate the state of multiple nonstationary processes is described. The concept of a constrained channel is introduced as a channel that is not capable of transferring all of the information required. A measure of information is developed based on the estimation entropy utilizing the Kalman filter state estimator. It is shown that this measure of information can be used to determine which process to observe in order to maximize a measure of global information flow. For stationary processes, the sampling sequence can be computed a priori, but nonstationary processes require real-time sequence computation.>

The authors present a pipeline method for detection and tracking of pixel-sized moving targets with unknown trajectories from a time sequence of highly noisy images. The pipeline target detection algorithm uses the temporal continuity of the smooth trajectories of moving targets and successfully detects and simultaneously tracks all the target trajectories by mapping them from the image sequence onto a single target frame. The pipeline method overcomes the constraint of a straight line trajectory that most other algorithms require for similar tasks. The algorithm is a complete parallel distributed processing type process, and therefore is highly time-efficient-ideal for real-time detection and tracking of arbitrary target trajectories in high-noise environments.>

A pattern-recognition system that self-organizes to recognize objects by shape as part of an integrated visual network (IVN) for autonomous flight control is presented. The system uses a multistaged hierarchical neural network that exhibits insensitivity to the location of the object in the visual field. The network's two layers perform the functionally disjoint tasks of invariance (position normalization) and recognition (identification of the shape). The invariance stage is a multilayered neural network implementation of a modified Walsh-Hadamard transform that generates a representation of the object that is invariant with respect to the object's position. The recognition stage is a modified version of the Fukushima neocognitron that identifies the position-normalized representation by shape. The inclusion of the invariance stage allows reduction of the massively replicated processing structures used for translation invariance in the neocognition. This system offers roughly the same translation invariance capabilities as the neocognitron, with a dramatic reduction in the number of elements and the network's interconnection complexity

This paper presents a pattern recognition system that self-organizes to recognize objects by shape as part of an Integrated Visual Network (IVN) for autonomous flight control. The system uses a multistaged hierarchical neural network that exhibits insensitivity to the location of the object in the visual field. The network's three layers perform the functionally disjoint tasks of preprocessing (dynamic thresholding), invariance (position normalization), and recognition (identification of the shape). The Preprocessing stage uses a single layer of elements to dynamically threshold the grey level input image into a binary image. The Invariance stage is a multilayered neural network implementation of a modified Walsh-Hadamard transform that generates a representation of the object that is invariant with respect to the object’s position. The Recognition stage is a modified version of Fukushima's Neocognitron that identifies the position normalized representation by shape. The inclusion of the Preprocessing and Invariance stages allows reduction of the massively replicated processing structures used for translation invariance in the Neocognitron. This system offers roughly the same translation invariance capabilities as the Neocognitron with a dramatic reduction in the number of elements and the network's interconnection complexity.

A fast tracking algorithm is used to determine displacements of points (pixels) on object surfaces between frames, a sequence of image frames representing a time-varying scene. The displacement is represented by a four-dimension vector whose elements are two-dimensional translation, rotation, rotational angle on the x-y plane, and scaling factor (zoom parameter). The algorithm was developed in conjunction with research on sensors for machine vision with biological vision features. The proposed sensor has two configurations: a uniform rectangular grid and a nonuniform logarithmic spiral grid. The rectangular grid is very well suited to translational motion in 2-D, whereas the logarithmic spiral grid has been proven to be appropriate for rotation and scaling. The algorithm developed utilizes both configurations for estimating 3-D motion.>

An investigation of the effects of various error components in a mobile robot system, which uses a navigation line, was previously reported by the author in 1985. An analysis of the effects of these error components on the navigation line-width determination and obstacle-range determination is presented that provided equations that are used to predict the error resulting from changes in various system parameters. In particular, the change in the rotation angle of the robot sensor about its optical axis, called the roll angle, is examined. Numerical examples showing the characteristics of the derived error equations are also given.>

A video tracking system based on the principles of optimalizing control is presented. This system is designed such that expensive segmentation procedures are not necessary, and is well suited for fast hardware implementation. At the image received from the camera a perturbation procedure is performed, simulating the variations on the intensity values that would occur if the camera were moved. An index of performance (IP) is evaluated at the results of this perturbation; this IP does not need to detect the position of the tracked object accurately, but rather varies in such a way as to detect the motion of the object with respect to the camera. Algorithms for translation and scaling of the tracking object have been developed using area, centroid, and template matching as indices of performance. Software simulation experiments show that the control law developed yields good results even under variations in background and illumination and under multidimensional motion.>

An algorithm for translational motion prediction is presented. The algorithm predicts motion without using correspondence and using only one line correlation instead of full image correlation. The algorithm was developed in conjunction with research on sensors for machine vision with biological vision features. Tests with noisy synthetic images produced good results.

Experimental results are presented that support theory published in the literature concerning the use of a navigation line for the guidance of mobile robots. A method for the determination of a robot's position is developed. A specialized edge operator, which aids in the segmentation of a navigation line from an image of a robot's environment, is given. Use of this specialized edge operator in conjunction with the Hough transform is also presented. These methods are used to verify the analytical results given previously. Comparisons are made between experimental data and expected results as a function of various system parameters. Real-time implementation of these methods is considered.

A continuous, straight-edged line is used for the visual navigation of an autonomous mobile robot in a factor environment. This line, which resides on the floor and contrasts with background, may also be used to determine range information. Two methods are developed for determining the range of an object in the sensor's field of view. The effects of various error conditions in the system geometry on each ranging method are determined. Equations are derived which yield the percent error in calculating ranges given estimates of these error conditions. Numerical examples using typical sensor parameters are given.

An image processor used as a position transducer is treated on an input-output basis. If processing speed is sufficiently fast the input-output characteristic is independent of frequency and can be represented as a gain function. The input is a function of the image pattern (including intensity if grey level is used) and the output is used for control or other purposes but control is the issue considered here. A Fourier phase in conjunction with projection concepts is used to obtain control signals in the image processing. A solution to two-dimensional rotation of the object input pattern is presented. The technique can easily be extended to deal with dilation problem. Results that are important to closed loop design are included.

The use of a contrasting line for the visual navigation of autonomous mobile robots in a factory environment is developed. Minimum and maximum linewidths are determined analytically by considering sensor geometry, field of view, and error conditions present in the system. The effects of these error conditions on the width of the line, as seen in the image plane, determines the optimal linewidth. Numerical examples using typical sensor parameters are given.

Research on the semiautonomous operation of mobile robots in typical pathways is described. The image of the pathway will consist of two nearly vertical lines bounding a region with little texture (the pathway) after correction for perspective. In order to identify pathway boundaries, regions in the image space are examined using an edge detection algorithm, edges between regions are determined by the algorithm, and those corresponding to straight or nearly straight lines with large slope (path boundaries) are identified by means of the Hough transform. Once the path boundaries are identified, the horizontal distance from camera to road edge is determined. Next, a method to detect cubics in the roadway (i.e., obstacles) is presented. The region of interest in the roadway (from the camera to some predetermined distance in front of it) is known from the path boundary algorithm. The interior of this region is examined for edges. If edges are detected, it means that obstacles or shadows are present. A method to separate obstacles from shadows using stere vision is then presented.

The distance between an object and stereo vision sensors can be measured using image processing and known system parameters. A detailed distance measurement synthesis procedure to meet system specifications is presented and illustrated with an example. An error analysis shows that error is proportional to distance. System parameters such as separation between sensor elements, sensor focal length, and sensor array dimensions are related in the design and error equations presented. The main desired design goal is to establish the smallest image sensor array size which will meet system operating specifications. Minimum and maximum distance, object height, optic parameters, scene shift, and sensor array parameters are related.

Algorithms are developed for a real-time, automatic system capable of tracking two-dimensional (2-D) targets in complex scenes. A mathematical model of 2-D image spatial and temporal evolution applicable to certain classes of targets and target perturbations is developed. It is shown that for small target perturbations the 2-D tracking problem may be approximated as a 1-D time-varying parameter estimation problem. A CCD-implementable solution is developed which is particularly useful when considering target rotation and dilation and target/background separation. Results of system simulation and suggestions for future research are presented.

Implementation of a three-dimensional video tracking algorithm using CCD's is presented. The algorithm does not require conventional pattern recognition or continuous segmentation updating. The only requirements, for the class of targets presented, are that the sampling lattice contain only points inside the target and that the background vary slowly. Several well-known segmentation techniques are available for initial segmentation and determination of target points. The use of CCD devices simplifies the architecture of the processor and permits real time processing at a high sampling rate. Some of the vector and matrix operations require either the use of continuously variable tapweight transversal filters or analog multipliers. The use of analog multipliers, which can be on-chip, reduces the number of components and offers higher accuracy than presently available variable tap weight transversal filters.

The development of algorithms for real-time, automatic systems capable of tracking 2-D targets in complex scenes is discussed. The problems and constraints involved in such an undertaking as well as previous efforts are summarized. in addition, a mathematical model of scene spatial and temporal evolution is developed for certain classes of targets and target perturbations. It is shown that for small target perturbations the 2-D tracking problem may be approximated as a 1-D time-varying parameter estimation problem. Algorithms resulting from several estimation approaches are discussed.

The use of charge-coupled devices (CCD's) for pattern recognition applications is discussed. In particular, it is suggested that the new CCD technology may satisfy the large memory, fast processing speed, low cost, small size, and low power requirements which could make practical many pattern processing systems which until now have not been feasible. Specific implementations are suggested along with estimates of processing times.

A method is presented which uses easily calculated probabilities to determine complete covers of prime implicant charts. Apparently the underlying principle of incomplete branching has not been applied previously to prime implicant charts. This principle differs in that branching is not carried out to complete covers, but only to a specified level, at which point each position is evaluated. All positions are then mimimaxed to the point of branching to choose the best branch. The method is easy to program and requires relatively little computer time to determine the cover. Time savings of up to 98 percent have been realized with an increase in the cost of the cover of less than 2 percent over conventional minimization methods. The method can be applied directly to any prime implicant chart, or can be used as a substitute for complete branching in charts where the application of dominance produces a cyclic chart.