China
Africa
Explore chapters and articles related to this topic
Automatic Sheep Age Estimation Based on Active Contours without Edges
Published in Siddhartha Bhattacharyya, Václav Snášel, Indrajit Pan, Debashis De, Hybrid Computational Intelligence, 2019
Aya Abdelhady, Aboul Ella Hassanien, Aboul Aly Fahmy
Blobs stand for Binary Large Objects. Blobs are the connected areas in white and black images [27]. Therefore, RGB images should be first converted into black and white images to find the connected components in a chain. From any point in the blob, neighbors are checked and tracked in all directions, north, east, west, south, northeast, southeast, southwest, northwest. Then each detected blob is given a unique label for counting. Moreover, centroids of the detected objects are determined to be able to measure a variety of blob properties.
Digital Image Processing for Machine Vision Applications
Published in Sheila Anand, L. Priya, A Guide for Machine Vision in Quality Control, 2019
Blob stands for “binary large object.” The method of analyzing an image which has undergone binarization processing is called “blob analysis.” Blob analysis is one of the basic methods used for analyzing the shape of an object. In blob analysis, we first separate the different objects in an image and then try to evaluate which object we are looking to recognize. For example, the objective may be to look for circles, squares, or other shapes present in a target image.
Automated method for detecting and reading seven-segment digits from images of blood glucose metres and blood pressure monitors
Published in Journal of Medical Engineering & Technology, 2019
E. Finnegan, M. Villarroel, C. Velardo, L. Tarassenko
In order to separate the blobs that corresponded to digit segments from those that are noise, a filtering algorithm was developed. This filtering algorithm is a combination of rule-based filtering and classification on several extracted features. A database of blobs was created by running the blob extraction algorithm on all images in the medical device image dataset. The blobs were saved as a 52 × 52 pixel binary image and manually labelled by the authors as either segments or noise. The medical device image dataset was split into training and test sets. The blobs extracted from the training sets were used to develop the filtering algorithm and the blobs from the test sets were used to test its performance.
Image feature detection using an improved implementation of maximally stable extremal regions for augmented reality applications
Published in International Journal of Image and Data Fusion, 2018
Neetika Gupta, Mukesh Kumar Rohil
In an attempt to design an affine invariant feature detection algorithm, Alvarez and Morales (1997) introduced an affine morphological multi-scale analysis to extract corners in an image. Tuytelaars and Gool (1999, 2000) proposed two approaches for detecting image features in an affine invariant way. The former approach extracted Harris points and used the nearby edge for defining a parallelogram region. The latter approach initiated by extracting of local intensity extrema and an ellipse was defined for the region determined by significant changes in the intensity profiles. Laptev and Lindeberg (2003) developed a method for finding elliptical blobs in an image for hand tracking.
Recognition of dynamic objects from UGVs using Interconnected Neural network-based Computer Vision system
Published in Automatika, 2022
In real-time motion, some processes are run sequentially. With the MOD process, moving objects are detected as blobs on the image. The MOR process starts immediately after this process. In this process, the real object and noise blobs detected in the previous step are first classified as objects or noise, and then the object recognition process is made by determining the classes of the objects. The UGV used in the study, shown in Figure 6, has two 420W DC motors and two BTS 7960 PN motor drivers. 1024 ppr encoder mounted on both wheels.