Bhoopalan Ramasubramanian

Detection of moving object from a visual sequence plays a vital role for the tracking of object. The main objective of this proposed work is to detect and classify the various video sequences with the help of different classification algorithms. The input video sequences from the publicly available datasets are collected and the individual frames are extracted. These frames are pre-processed and then applied to the novel background subtraction process. Important features based on the Local Binary Pattern (LBP) and grey level co-efficient are extracted. Finally these features are classified by three different classifiers like SVM, PLS, and PNN. The performance of these different classifiers are evaluated and compared. It is found that PLS classifier produces more classification accuracy but with more computation time.

Recently, researchers have built new deep learning (DL) models using a single image modality to diagnose age-related macular degeneration (AMD). Retinal fundus and optical coherence tomography (OCT) images in clinical settings are the most important modalities investigating AMD. Whether concomitant use of fundus and OCT data in DL technique is beneficial has not been so clearly identified. This experimental analysis used OCT and fundus image data of postmortems from the Project Macula. The DL based on OCT, fundus, and combination of OCT and fundus were invented to diagnose AMD. These models consisted of pre-trained VGG-19 and transfer learning using random forest. Following the data augmentation and training process, the DL using OCT alone showed diagnostic efficiency with area under the curve (AUC) of 0.906 (95% confidence interval, 0.891–0.921) and 82.6% (81.0–84.3%) accuracy rate. The DL using fundus alone exhibited AUC of 0.914 (0.900–0.928) and 83.5% (81.8–85.0%) accuracy rate. Combined usage of the fundus with OCT increased the diagnostic power with AUC of 0.969 (0.956–0.979) and 90.5% (89.2–91.8%) accuracy rate. The Delong test showed that the DL using both OCT and fundus data outperformed the DL using OCT alone (P value < 0.001) and fundus image alone (P value < 0.001). This multimodal random forest model showed even better performance than a restricted Boltzmann machine (P value = 0.002) and deep belief network algorithms (P value = 0.042). According to Duncan’s multiple range test, the multimodal methods significantly improved the performance obtained by the single-modal methods. In this preliminary study, a multimodal DL algorithm based on the combination of OCT and fundus image raised the diagnostic accuracy compared to this data alone. Future diagnostic DL needs to adopt the multimodal process to combine various types of imaging for a more precise AMD diagnosis.

Advances in the eye care telemedicine system aid the diabetic patients in remote areas to stop the unwanted visit to ophthalmologist, reduces overall cost, time and money. Diabetic retinopathy, which is the primary cause of sight loss, has the most common symptoms like microaneurysms, haemorrhages, cotton‐wool spots, exudates and drusen. In this work, an efficient approach for the automatic detection of haemorrhages in colour retinal images is proposed and validated. The colour retinal images captured from the diabetic patients are enhanced by an effective pre‐processor. A bag of features based on intensity, colour and texture are extracted. Finally, the features are classified with the help of partial least square classifier. The classifier performance is validated on two publicly available databanks. The developed method obtains an area under receiver operating characteristic curve of 0.98 with an average execution time of 6 s. This application outperforms the existing approaches with high robustness and efficiency.

Diabetic Retinopathy (DR) is the primary cause of visual loss in case of diabetic patients. It is mainly caused by the changes in the blood vessels of the retina even though manual screening is available but they are very much time consuming and inefficient. Therefore for automatic detection of these changes and prevention of vision loss, color retinal image is an important tool used by the ophthalmologist. This paper discusses on seven different preprocessing available in literature and found the optimum technique for enhancing the retinal images. The performance of all the techniques has been tested on more than 1000 images obtained from several publicly available DRIVE and STARE and MESSIDOR dataset.

Glaucoma and Diabetic Retinopathy are the leading common cause of vision loss. Optic Disc and macula are the important landmark for the detection of these retinal pathology. Even though, the manual screening of Optic Disc and macula are available, they consume more time and have more human error. Therefore, there is a need of an automated application for the reliable and efficient localization of OD and macula. This paper presents a novel and fast stand-alone MATLAB application for the segmentation of these anatomical structures. This application uses colour K-means segmentation for locating macula and for detecting Optic Disc, bilateral filtering followed by Morphological operation is used. The application were evaluated on 135 glaucomatous images collected from local eye hospital and 200 diabetic retinopathy images from DIARETDB1 and DRIVE database. This stand-alone application consumes an average computation time of 13s for macula localization and 32s for OD segmentation with a success rate of 98%. Overall, the proposed application obtain an efficient result in the segmentation of both OD and macula, which in turn helps to detect glaucoma and Diabetic Retinopathy.