Camera distortion self-calibration using the plumb-line constraint and minimal Hough entropy

Abstract

In this paper we present a simple and robust method for self-correction of camera distortion using single images of scenes which contain straight lines. Since the most common distortion can be modelled as radial distortion, we illustrate the method using the Harris radial distortion model, but the method is applicable to any distortion model. The method is based on transforming the edgels of the distorted image to a 1-D angular Hough space, and optimizing the distortion correction parameters which minimize the entropy of the corresponding normalized histogram. Properly corrected imagery will have fewer curved lines, and therefore less spread in Hough space. Since the method does not rely on any image structure beyond the existence of edgels sharing some common orientations and does not use edge fitting, it is applicable to a wide variety of image types. For instance, it can be applied equally well to images of texture with weak but dominant orientations, or images with strong vanishing points. Finally, the method is performed on both synthetic and real data revealing that it is particularly robust to noise.

Keeping a Pan-Tilt-Zoom Camera Calibrated

“…Therefore, in this section, we use two images from the set PT that have the same tilt angle but different pan angles to estimate the important parameter ð0Þ, the lens distortion coefficient at the lowest scale. When long, straight lines are present in the scene, radial distortion can be estimated (e.g., [19]); however, we do not wish to impose any restrictions on scene content. Here, we propose a novel approach inspired by Fitzgibbon [7], who introduced a method for linearly estimating divisionmodel lens distortion during the estimation of the fundamental matrix for a moving camera.…”

Correcting radial and perspective distortion by using face shape information

“…We can roughly classify the previous works from the features that are used. Most of these methods used a structured pattern such as chessboard [1], dots, straight lines [11], [14], [15], [6], [12], or reference objects with known 3D objects. The methods using straight lines could be used in manmade environments because buildings usually contain a lot of straight lines.…”

Demisting the Hough Transform for 3D Shape Recognition and Registration

“…A similar approach, but minimizing entropy w.r.t. some parameters of the vote generation process, has already been used for lens distortion calibration [21]. A lower entropy distribution contains less information, making it more peaky and hence having more votes in agreement.…”