Automatic Control of Horizontal-moving Stereoscopic Camera by Disparity Compensation

Vergence Control of Binocular Stereoscopic Camera Using Disparity Information

“…2, extracts the horizontal disparity of key-object from the acquired left & right image, and minimizes the difference between the distance of a key-object and the convergence distance of camera by keeping disparity value to the minimum [1].…”

“…Parallel mode, however, is a binocular stereoscopic camera method of controlling vergence by horizontally moving separated lens and CCD sensor into symmetric position to each other. Either of them can easily acquire the quantity of vergence control θ and h by interpreting their own geometric structure [1][2][3]. With the distance between lens and key-object as p, the distance between CCD sensor and lens as f, and the distance between left and right camera as 2t, the vergence control angle of toed-in camera (θ) and the control quantity of parallel camera (h) can be expressed as follows: Vergence Control of Binocular Stereoscopic Camera…”

“…Human beings repeat focus control of the thickness of the eye lens, according to the distance, and vergence control, which always keeps the key-object formed in the center of both the retinas and so, provides evernatural and clear image information [1][2][3][4][5]. The multiple cameras with two or more sensors or a lens array composed of many lenses are used to get three-dimensional data [6][7][8].…”

“…Coombs, etc. proposed the real-time vergence control system using a Cepstral filter for estimating the quantity of vergence control error in the acquired left & right stereoscopic images. [12][13][14] Last, K. C. Kwon [1] showed the theory of relationship between vergence and focus according to the object distance of parallel stereoscopic cameras, and took advantage of one-dimensional Cepstral Filter with the vertical projection data of a pair of stereoscopic images as input data for fast, accurate simultaneous control of vergence and focus. The methods of extracting disparity information that have been proposed for the vergence control of stereoscopic cameras have long-time processing and low accuracy, and also most of them are vergence control methods of toed-in stereoscopic cameras.…”

Precision Evaluation of Three-dimensional Feature Points Measurement by Binocular Vision

“…Visual navigation, robotics and vision-based measurement are just a few examples of industrial applications that depend on pose estimation and computation of the three-dimensional (3D) object location [1][2][3][4][5][6][7]. In 3D reconstruction process with an optical technique, system accuracy is defined as a problem of the determination of 3D point positions, which are provided by 2D features gathered from binocular calibrated cameras.…”