China
Africa
Explore chapters and articles related to this topic
Fundamentals of image analysis and interpretation
Published in Michael O’Byrne, Bidisha Ghosh, Franck Schoefs, Vikram Pakrashi, Image-Based Damage Assessment for Underwater Inspections, 2019
Bidisha Ghosh, Michael O’Byrne, Franck Schoefs, Vikram Pakrashi
This section deals with: Point operations—these operations deal with one pixel at a time, with no consideration of neighboring pixels. Other point processing examples include color conversions and numeric data conversions.Neighborhood operations—these operations consider the local neighborhood around the processed pixel when computing the output. Examples include filtering-based operations and morphological operations.Image restoration/enhancement methods using multiple images—these operations rely on data from multiple images to create one “super” image that contains more information than any of the constituent images.Geometric transformations—these operations modify the spatial relationships between pixels in an image. They are often used for image alignment and correcting for lens distortion.
Signal Recovery from Partial Information
Published in Vijay K. Madisetti, The Digital Signal Processing Handbook, 2017
The simple signal degradation model described in the next section turns out to be a useful representation for many different problems encountered in practice. Some examples that can be formulated using the general signal recovery paradigm include image restoration, image reconstruction, spectral estimation, and filter design. We distinguish between image restoration, which pertains to image recovery based on a measured distorted version of the original image, and image reconstruction, which refers most commonly to medical imaging where the image is reconstructed from a set of indirect measurements, usually projections. For many of the signal recovery applications, it is desirable to extrapolate a signal outside of a known interval. Extrapolating a signal in the spatial or temporal domain could result in improved spectral resolution and applies to such problems as power spectrum estimation, radio astronomy, radar target detection, and geophysical exploration. The dual problem, extrapolating the signal in the frequency domain, also known as superresolution, results in improved spatial or temporal resolution and is desirable in many image restoration problems. As will be shown later, the standard inverse filtering techniques are not able to resolve the signal estimate beyond the diffraction limit imposed by the physical measuring device.
Digital Image Processing and Analysis
Published in Scott E. Umbaugh, Digital Image Processing and Analysis, 2017
Image restoration is the process of taking an image with some known, or estimated, degradation, and restoring it to its original appearance. For example, image restoration is used in the field of law enforcement where an image has been degraded by blurring due to motion. We may need to read the letters on a blurred license plate from a moving car or identity a face in the car. We need to model the blurring process to develop a model for the distortion. Once we have a model for the degradation process, we can apply the inverse process to the image to restore it to its original form. Image restoration is often used in space exploration—for example, to eliminate artifacts generated by mechanical jitter in a spacecraft (Figure 1.3-1) or to compensate for flaws in the optical system of a telescope. Restoration techniques can be used in noise removal, as shown in Figure 1.3-2, or in fixing geometric distortion, as shown in Figure 1.3-3.
Integrated Recognition Assistant Framework Based on Deep Learning for Autonomous Driving: Human-Like Restoring Damaged Road Sign Information
Published in International Journal of Human–Computer Interaction, 2023
Jeongeun Park, Kisu Lee, Ha Young Kim
Image restoration includes a large range of detailed tasks, such as image inpainting and enhancement. Deep learning-based image restoration tasks include filling in a portion of a damaged image using surrounding pixel information or removing an object from the entire image (Su et al., 2022). Unsupervised learning-based deep learning models have steadily been developed for image restoration. This study focuses on restoring an image obstructed from view by another object to the uncovered image. Image restoration aims to restore original images to be visible to humans. The method of applying deep learning to image restoration has received great attention, as it could exhibit higher performance than traditional image restoration methods. Deep learning algorithms for image restoration are primarily configured based on deep convolutional neural networks (DCNNs). In particular, the corresponding algorithms may be classified into algorithms that use a generative adversarial network (GAN) structure and those that do not (Goodfellow et al., 2020; Su et al., 2022).
On image restoration from random sampling noisy frequency data with regularization
Published in Inverse Problems in Science and Engineering, 2019
Image restoration can be roughly divided into three different kinds of problems, namely, image denoising, image deblurring and image enhancement, with the main purpose of recovering a clear image from its contaminated measurement data. There are two main issues to be dealt with during the imaging process. Firstly, the given noisy measurement data of the image may be incomplete, leading to non-unique reconstructions of the desired image in principle. In this case, we can only find some approximate reconstructions using insufficient noisy data. Secondly, since the salient ingredients of an image are its edges, the discontinuities of the grey function of the image should be kept, when some denoising process is applied to remove the contamination of the image.
Least-squares solutions of the reduced biquaternion matrix equation AX=B and their applications in colour image restoration
Published in Journal of Modern Optics, 2019
The field of image restoration is required to retrieve the information from degraded images. The purpose of image restoration is the removal or reduction of degradations which are included during the acquisition of images, e.g. noise, pixel value errors, out of focus blurring or camera motion blurring using prior knowledge of the degradation phenomenon. This means it deals with the modelling of the degradation and applying the process (inverse) to reconstruct the image (see Figure 2)(12, 13). The image restoration has got a wide scope of usage e.g. scientific explorations, legal investigations, film making and archival, image and video (de-)coding, consumer photography, etc.