Light microscopic iris classification using ensemble multi‐class support vector machine

Abstract

Similar to other biometric systems such as fingerprint, face, DNA, iris classification could assist law enforcement agencies in identifying humans. Iris classification technology helps law‐enforcement agencies to recognize humans by matching their iris with iris data sets. However, iris classification is challenging in the real environment due to its invertible and complex texture variations in the human iris. Accordingly, this article presents an improved Oriented FAST and Rotated BRIEF with Bag‐of‐Words model to extract distinct and robust features from the iris image, followed by ensemble multi‐class‐SVM to classify iris. The proposed methodology consists of four main steps; first, iris image normalization and enhancement; second, localizing iris region; third, iris feature extraction; finally, iris classification using ensemble multi‐class support vector machine. For preprocessing of input images, histogram equalization, Gaussian mask and median filters are applied. The proposed technique is tested on two benchmark databases, that is, CASIA‐v1 and iris image database, and achieved higher accuracy than other existing techniques reported in state of the art.

Machine learning techniques to detect and forecast the daily total COVID‐19 infected and deaths cases under different lockdown types

“…The planes that separate the classes into higher dimension are called hyperplanes (Nobel, 2006). The SVR training algorithms are mostly offline, but online algorithms are mostly used to automatically track system model time-varying changes and time lagging F I G U R E 4 Flowchart to forecast the daily cumulative infected and death cases due to COVID-19 characteristics Rehman (2021). The online SVR algorithms have drawbacks like when the margin support vector is empty and the training speed is plodding.…”

Deep convolutional generative adversarial network for Alzheimer's disease classification using positron emission tomography (PET) and synthetic data augmentation

“…In model‐based techniques, the data is observed, a model is developed, and a dedicated engine is used to extract the data. This technique has been utilized to increase the training data similar to an urban driving environment (Richter et al, 2016), object detection (Aurangzeb et al, 2019; Khan et al, 2019b), forecasting (Phetchanchai et al, 2010; Saba, Rehman, & AlGhamdi, 2017), realistic digital brain image generation (Khan et al, 2021), text data segmentation Rehman, 2021; Harouni et al, 2014; Rehman, Kurniawan, & Saba, 2011; Saba & Rehman, 2012; Saba, Rehman, Al‐Dhelaan, & Al‐Rodhaan, 2014; Rehman et al, 2011), and synthetic agar plate image generation (Khan et al, 2021; Husham, Alkawaz, Saba, Rehman, & Alghamdi, 2016; Nodehi et al, 2014). However, to design such an accurate and dedicated data generation engine, deep knowledge with the desired domain expertise is essential.…”

Computer-aided analysis of radiological images for cancer diagnosis: performance analysis on benchmark datasets, challenges, and directions

“…Compared to their initial counterparts, secondary tumour extension into the lumens of the abdominal veins seems more prevalent. The proclivity for venous invasion is well documented in tumours from the liver and kidneys; nevertheless, various other malignancies may also extend into the venous lumen (Rehman 2020 , 2021 ).…”