Two-dimensional image motion is the projection of the three-dimensional motion of objects, relative to a visual sensor, onto its image plane. Sequences of time-orderedimages allow the estimation of projected two-dimensional image motion as either instantaneous image velocities or discrete image displacements. These are usually called the optical flow field or the image velocity field. Provided that optical flow is a reliable approximation to two-dimensional image motion, it may then be used to recover the three-dimensional motion of the visual sensor (to within a scale factor) and the three-dimensional surface structure (shape or relative depth) through assumptions concerning the structure of the optical flow field, the three-dimensional environment, and the motion of the sensor. Optical flow may also be used to perform motion detection, object segmentation, time-to-collision and focus of expansion calculations, motion compensated encoding, and stereo disparity measurement. We investigate the computation of optical flow in this survey: widely known methods for estimating optical flow are classified and examined by scrutinizing the hypothesis and assumptions they use. The survey concludes with a discussion of current research issues.

/pdf/the-computation-of-optical-flow-2n5afvnh1s.pdf

The computation of optical flow

An imaging system for a vehicle includes an imaging device having a field of view exteriorly and forward of the vehicle in its direction of travel, and an image processor operable to process the captured images in accordance with an algorithm. The algorithm comprises a sign recognition routine and a character recognition routine. The image processor processes the image data captured by the imaging device to detect signs in the field of view of the imaging device and applies the sign recognition routine to determine a sign type of the detected sign. The image processor is operable to apply the character recognition routine to the image data to determine information on the detected sign. The image processor applies the character recognition routine to the captured images in response to an output of the sign recognition routine being indicative of the detected sign being a sign type of interest.

Imaging system for vehicle

https://www.cns.nyu.edu/pub/lcv/simoncelli96.pdf

A model of neuronal responses in visual area MT.

A method of using stereo vision to interface with a computer is provided. The method includes capturing a stereo image, and processing the stereo image to determine position information of an object in the stereo image. The object is controlled by a user. The method also includes communicating the position information to the computer to allow the user to interact with a computer application.

Video-based image control system

A vehicle headlamp control method and apparatus includes providing an imaging sensor that senses light in spatially separated regions of a field of view forward of the vehicle. Light levels sensed in individual regions of the field of view are evaluated in order to identify light sources of interest, such as oncoming headlights and leading taillights. The vehicle's headlights are controlled in response to identifying such particular light sources or absence of such light sources. Spectral signatures of light sources may be examined in order to determine if the spectral signature matches that of particular light sources such as the spectral signatures of headlights or taillights. Sensed light levels may also be evaluated for their spatial distribution in order to identify light sources of interest.

Vehicle headlight control using imaging sensor

A low cost, real time vehicle lane position detection system is provided for determining and maintaining the position of the vehicle on a highway. The system comprises an image sensor mounted on the front of an automotive vehicle and an integrated processor for performing real-time lane mark detection. The sensor/processor system identifies highway lane marks on the detector image plane by using a nonlinear resistive network for detecting outliers. A line detection algorithm, such as the Hough transform, is used to determine the lane marks from the outliers on the image plane. Because the expected lane position can be determined in advance, an added degree of signal-to-noise discrimination is achieved by providing feedback to the processor for outlier detection. The position of the vehicle in the lane is determined and tracked from the position of the detected lane marks on the image plane given the sensor position and optical geometry.

Vehicle lane position detection system

Analog circuits based on resistive networks emulate the behavior of the vertebrate eye, detecting edges, differentiating between surfaces, and estimating motion. The speed and simplicity of these devices make them candidates for applications ranging from security systems to retinal replacements.

Neuromorphic vision chips

A system of active emitters and sensors is provided for measuring distance and communicating between vehicles on an automated highway. The preferred system utilizes a pair of spaced apart sensors at the front of each vehicle for stereo depth perception, two temporally modulated emitters on the rear of each vehicle for redundancy and inter-vehicle communication, and temporally modulated emitters positioned at intervals along the highway for communication from the highway to the vehicles. The emitters may transmit radiation at a wavelength, such as 1 μm IR, for example, that can be detected by low cost detectors. The emitters are modulated temporally to transmit a binary code, resulting in signal-to-noise improvement and increased clutter rejection and operational range. Each sensor may include an array of IR detectors and associated readout circuitry, an imaging lens, a lens array for improving detector fill factor, a filter and polarizer to reject clutter and reduce noise, and a cylindrical lens to provide a good vertical field of view. Detector charge integration is controlled by the binary code of the emitters and depends on the code length and local background illumination. The sensors are connected to a computer processor for processing data received from the emitters and computing distance between vehicles.

Inter-vehicle distance measuring system

Computing motion on the basis of the time-varying image intensity is a difficult problem for both artificial and biological vision systems. We show how gradient models, a well-known class of motion algorithms, can be implemented within the magnocellular pathway of the primate's visual system. Our cooperative algorithm computes optical flow in two steps. In the first stage, assumed to be located in primary visual cortex, local motion is measured while spatial integration occurs in the second stage, assumed to be located in the middle temporal area (MT). The final optical flow is extracted in this second stage using population coding, such that the velocity is represented by the vector sum of neurons coding for motion in different directions. Our theory, relating the single-cell to the perceptual level, accounts for a number of psychophysical and electrophysiological observations and illusions.

/pdf/computing-optical-flow-in-the-primate-visual-system-3mu5tftezu.pdf

Bimal P. Mathur

Papers

Vehicle lane position detection system

Neuromorphic vision chips

Inter-vehicle distance measuring system

Computing optical flow in the primate visual system

Visual Information Processing: From Neurons to Chips