Vignetting refers to the fall-off pixel intensity from the centre towards the edges of the image. This effect is undesirable in image processing and analysis. In the literature, the most commonly used methods of vignetting correction assume radial characteristic of vignetting. In the case of camera lens systems with non-radial vignetting, such approach leads to insufficient correction. Additionally, the majority of vignetting correction methods need a reference image acquired from a uniformly illuminated scene, what can be difficult to achieve. In this paper, we propose a new method of vignetting correction based on the local parabolic model of non-radial vignetting and compensation of non-uniformity of scene luminance. The new method was tested on camera lens system with non-radial vignetting and non-uniformly illuminated scene. In these conditions, the proposed method gave the best correction results among the tested methods.

This article concerns the issue of image semantic segmentation for the machine vision system of an autonomous Unmanned Ground Vehicle (UGV) moving in an off-road environment. Determining the meaning (semantics) of the areas visible in the recorded image provides a complete understanding of the scene surrounding the autonomous vehicle. It is crucial for the correct determination of a passable route. Nowadays, semantic segmentation is generally solved using convolutional neural networks (CNN), which can take an image as input and output the segmented image. However, proper training of the neural network requires the use of large amounts of data, which becomes problematic in the situation of low availability of large, dedicated image data sets that consider various off-road situations - driving on various types of roads, surrounded by diverse vegetation and in various weather and light conditions. This study introduces a synthetic image dataset called “OffRoadSynth” to address the training data scarcity for off-road scenarios. It has been shown that pre-training the neural network on this synthetic dataset improves image segmentation accuracy compared to other methods, such as random network weight initialization or using larger, generic datasets. Results suggest that using a smaller but domain-dedicated set of synthetic images to initialize network weights for training on the target real-world dataset may be an effective approach to improving semantic segmentation results of images, including those from off-road environments.

The effect of vignetting is undesirable in image processing and analysis. It cause fall-off of pixel intensity from centre towards edges of the image. In this paper, we propose a new procedure of fast vignetting reduction based on two images acquired with different camera/lens settings. The change of lens aperture or focal length will also change the effect of vignetting in images. These differences are used to estimate a vignetting function, which is used for vignetting reduction. The obtained image after vignetting reduction is not significantly different from the image without vignetting.

The vignetting is one of image distortions, which refers to the fall-off of pixel intensity due to optical properties of the camera and lens system. The effect of vignetting on image is unwanted in image processing and analysis. The vignetting of the camera-lens system can be determined using the vignetting model. The local vignetting polynomial model allows to fit the vignetting data with more flexibility, but at the cost of increased of vignetting function distortions. This paper introduces a new smooth local polynomial vignetting model, which is a modification of local vignetting polynomial model. The conducted experiments on broad range of lens settings shows that filtering allows to improve the vignetting function smoothness.

The procedure of camera color calibration ensures accurate and repeatable acquisition of scene colors. The main process of calibration procedure is the color transformation from camera RGB color space to device independent color space, such as CIEXYZ or CIELAB. Typical calibration procedure assumes the uniformity of irradiance across the whole scene, but such assumption is difficult to achieve. Spatial changes in irradiance are typical for indoor and outdoor light conditions. The problem of color calibration under non-uniform lighting was researched by, e.g., B. Funt and P. Bastani. In the present article, their calibration procedure was tested together with the classic approach to camera calibration. Based on these experiments a modification was proposed, in which RGB image values are scaled accordingly to the results of additional measurements. This modification allows to obtain, in conditions of spatially nonuniform scene illumination, lower values of color differences ΔE * ab and ΔE00 than in cases of classic method and FuntBastani method.