China 
Africa 
Explore chapters and articles related to this topic
Introduction to Digital Image Analysis
Published in Scott E. Umbaugh, Digital Image Processing and Analysis, 2017
Another term for image subtraction is background subtraction, as we are really simply removing the parts that are unchanged, the background. Although the process is the same as in motion detection, it is thought of differently. In comparing complex images, it may be difficult to see small changes. By subtracting out common background image information, the differences are more easily detectable. Medical imaging often uses this type of operation to allow the medical professional to more readily see changes which are helpful in the diagnosis. The technique is also used in law enforcement and military applications, for example, to find an individual in a crowd or to detect changes in a military installation. The complexity of the image analysis is greatly reduced when working with an image enhanced through this process.
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
Although image averaging is shown here, the same basic statistical principle can be applied to similar types of operations, such as finding the median value at each pixel location, which also serves to attenuate noise. Additionally, image subtraction finds many uses in image processing. It can be useful for the subtraction of a known pattern (or image) of superimposed noise or, indeed, for motion detection: stationary objects cancel each other out while moving objects are highlighted when two images of the same dynamic scene, which have been taken at slightly different times, are subtracted. This process of subtraction of an uninteresting background image from a foreground image containing information of interest is referred to as “background subtraction.”
Multiple Image Techniques
Published in Brian E. Dalrymple, E. Jill Smith, Forensic Digital Image Processing, 2018
Brian E. Dalrymple, E. Jill Smith
It is essential to the effectiveness of an image subtraction that when possible, the lighting, camera, and subject position remain unchanged in both images. Image registration must be preserved or acquired prior to subtraction. If the angle of ambient lighting were to be different in the second image, artifacts would be created, even if the registration is preserved. Ideally, the only difference between the two images should be the presence of the fingerprint, scale, and identifying markings in the first one. Auto-alignment of images within Photoshop can bring two misaligned images into registration, but it cannot solve the problem created by changes in lighting between the input images.
Optimised Feature Selection for Identification of Carcinogenic Leukocytes Using Weighted Aggregation Based Transposition PSO
Published in IETE Journal of Research, 2022
Subhajit Kar, Kaushik Das Sharma, Madhubanti Maitra
For each segmented leukocyte: Evaluate cluster index metric.Calculate cluster centre values.Sort cluster centre values in ascending order.Extract brightness values of each pixel from the cluster which contains the leukocyte component.Perform global threshold operation.Perform image subtraction operation between leukocyte and nucleus.
Haemorrhages detection using geometrical techniques
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2020
The green channel is used due to its high contrast. A preprocessing module consists of two steps: To correct non-uniform illumination through contrast enhancement and to enhance bright objects from the image. For later segmentation to be accurate for red lesion pathologies, switched median filtering is the preprocessing module to be applied on green channel image. For smoothing, progressive switched median filter, i.e. PSMF is used with the window size of 3. This filtering results in edge pixels preservation and noise blotches removal. Then, the normalisation through histogram stretch is applied on to get the resultant image as . To highlight bright objects from the overall image, subtraction is performed between and
An improved gray line profile method to inspect the warp–weft density of fabrics
Published in The Journal of The Textile Institute, 2019
Erdoğan Aldemir, Hakan Özdemir, Zekeriya Sarı
Gabor filtering has widely used in textile industry for the purpose of detection of fabric defects. A method has been proposed, particularly useful for patterned fabrics, that applies a Gabor filter tuned to match the texture information of nondefective fabric via the genetic algorithm to the fabrics to be detected (Jing, Yang, Li, & Kang, 2014). Jia & Liang (2017) proposed an automatic fabric defect inspection method based on lattice segmentation assisted by Gabor filtering. Two methods, one of them is based on Gabor filter and the other of them is based on the golden image subtraction, are proposed in Jing, Chen, and Li (2017) and compared to the wavelet pre-processed golden image subtraction method. In Bissi et al. (2013) proposed an algorithm to detect the defect in uniform and structured fabrics. The proposed algorithm comprises two phases, the first of which, also called as the feature extraction phase depends on complex Gabor filter bank and Principal Component Analysis, and the second of which, also called as the defect identification phase, is based on Euclidean norm of features and on the comparison with fabric type specification parameters. In addition to the all these aforementioned methods of Gabor filters in the detection of fabric defects, they will be used to identify the density of warp and weft yarn densities of the skewed fabrics.