Removal of specular reflections from image sequences using feature correspondences

Abstract

Abstract. The presence of specular highlights can hide underlying features of a scene within an image and can be problematic in many application scenarios. In particular, this poses a significant challenge for applications where image stitching is used to create a single static image of a scene from inspection footage of pipes, gas tubes, train tracks and concrete structures. Furthermore, they can hide small defects in the images causing them to be missed during inspection. We present a method which exploits additional information in neighbouring frames from video footage to reduce specularity from each frame. The technique first automatically determines frames which contain overlapping regions before the relationship that exists between them is exploited in order to suppress the effects of specular reflections. This results in an image that is free from specular highlights provided there is at least one frame present in the sequence where a given pixel is present in a diffuse form. The method is shown to work well on greyscale as well as colour images and effectively reduces specularity and significantly improves the quality of the stitched image, even in the presence of noise. While applied to the challenge of reducing specularity in inspection videos, the method improves upon the state-of-the-art in specularity removal, and, its applications are wideranging as a general purpose pre-processing tool.

Deep Highlight Removal Using Temporal Dark Prior in High-Speed Domain

“…There are two types of previous studies using multiple images: different directions of a light source [28], [29], [30] and different viewpoints [31], [32]. These methods focus on the spatial information of the images.…”

“…These methods focus on the spatial information of the images. References [28], [29], [30] used images taken from the same viewpoint with different illuminant positions. The datasets of [31] and [32] were obtained with different vantage points.…”

Mirror, Mirror, on the Wall, Who's Got the Clearest Image of Them All? - A Tailored Approach to Single Image Reflection Removal

“…A number of techniques have been developed which use information from multiple images to detect and remove reflec-tions. These include the use of different polarisation angles [5], [16], [6], [17], [3], adjustment of focal and flash settings [18], [1], [2], and the uses of relative motion and coherence [7], [8], [19], [20], [21], [22], [23], [24], [25]. A recent technique [26] seeks to improve on these methods by seeking to match the transmitted layer, while other techniques may erroneously match the reflected layer.…”

Obstacle Distance Measurement Under Varying Illumination Conditions Based on Monocular Vision Using a Cable Inspection Robot

“…These methods are roughly divided into four categories: multipleimage-based methods, single-image-based methods, learningbased methods and polarization filter methods. For multipleimage-based methods, Shah et al presented a specular highlight removal method from video frames by detecting correspondences [8]. Guo et al proposed a robust method to employ three prior structures of decomposed layers [9].…”