M
Majid Mirmehdi
Researcher at University of Bristol
Publications - 247
Citations - 6104
Majid Mirmehdi is an academic researcher from University of Bristol. The author has contributed to research in topics: Image segmentation & Active contour model. The author has an hindex of 38, co-authored 237 publications receiving 5523 citations. Previous affiliations of Majid Mirmehdi include City University London & Vision Institute.
Papers
More filters
Automatic Recognition of Exudative Maculopathy using Fuzzy C- Means Clustering and Neural Networks
TL;DR: An automatic method for the detection of exudate regions is introduced comprising image colour normalisation, enhancing the contrast between the objects and background, segmenting the colour retinal image into homogenous regions using Fuzzy C-Means clustering, and classifying the regions into exudates and nonExudates patches using a neural network.
Proceedings ArticleDOI
Detection and Tracking of Very Small Low Contrast Objects
TL;DR: A Kalman tracking algorithm that can track a number of very small, low contrast objects through an image sequence taken from a static camera using a combination of wavelet filtering for detection with an interest operator for testing multiple target hypotheses based within the framework of a Kalman tracker.
Journal ArticleDOI
Recognising text in real scenes
Paul Clark,Majid Mirmehdi +1 more
TL;DR: Two different approaches to the location and recovery of text in images of real scenes are presented, one using page edges and other rectangular boundaries around text, and the other using low-level texture measures with a neural network classifier.
Book ChapterDOI
Comparative Exudate Classification Using Support Vector Machines and Neural Networks
TL;DR: After segmenting candidate exudates regions in colour retinal images, the Neural Network based approach performs marginally better than the Support Vector Machine based approach, but it is shown that the latter are more flexible given criteria such as control of sensitivity and specificity rates.
Journal ArticleDOI
Recognizing Text-Based Traffic Signs
Jack Greenhalgh,Majid Mirmehdi +1 more
TL;DR: A novel system for the automatic detection and recognition of text in traffic signs using Maximally stable extremal regions and hue, saturation, and value color thresholding to locate a large number of candidates and interprets the text contained within detected candidate regions.