Namita Sengar

Automatic exudates detection in fundus image using intensity thresholding and morphology

Diabetic retinopathy (DR) is a leading cause of blindness in diabetic patients. Exudates are one of the most common earliest signs of diabetic retinopathy. Automatic and accurate detection of exudates in fundus images is an important step in early diagnosis of DR. In the proposed method detection of exudates, two independent approaches based on intensity thresholding and morphological processing are strategically combined to detect any small exudates present while removing all possible types of false positives. This strategic combination removes the noise sources from blood vessels and reflections during image capture making the detection of exudates accurate. Experimental results indicate that the proposed method has good accuracy in exudates detection without compromising the computational time and hence can be considered for screening purpose of DR.

Detection of diabetic macular edema in retinal images using a region based method

Diabetic macular edema (DME) is one of the severe complication of Diabetic Retinopathy. In this paper a method is proposed to detect the macula centre which is independent of optic disc location. Grading of DME is done by dividing the image of retina in different regions according to the international standard. Disease severity is accessed using scaling of bright lesions in macular regions. In this method search region for detection of macula is adaptive to the size of image. Independency from optic disc detection to detect the macula is an efficient method because it is unaffected by wrong detection of optic disc position under the presence of noises and reflections. The proposed method is tested on 100 images of MESSIDOR database and has achieved good accuracy 80 to 90 % for different cases.

Extraction of retinal vasculature by using morphology in fundus images

In this paper algorithm is proposed for detection of vessels present in a fundus image of an eye. Blood vessels extraction and removal are used to detect the other artifacts like lesions, the fovea and optic nerve. The proposed algorithm used the combination of different morphological operators which make this method less complex and also computationally efficient. Two different channels of an image green and L respectively are utilized to get the final vessel structure. This method also gives the region of interest for macula which may make macula detection easy. The proposed algorithm is tested on DRIVE data set of fundus image of an eye. The result gives good detection of vessel structure and the proposed method is computationally efficient.

Computer vision based technique for identification and quantification of powdery mildew disease in cherry leaves

AbstractThere are different reasons like pests, weeds, and diseases which are responsible for the loss of crop production. Identification and detection of different plant diseases is a difficult task in a large crop field and it also requires an expert manpower. In this paper, the proposed method uses adaptive intensity based thresholding for automatic segmentation of powdery mildew disease which makes this method invariant to image quality and noise. After the segmentation of powdery mildew disease from leaf images, the affected area is quantified which makes this method efficient for grading the level of disease infection. The proposed method is tested on the comprehensive dataset of leaf images of cherry crops, which achieved good accuracy of 99%. The experimental results indicate that proposed method for segmentation of powdery mildew disease affected area from leaf image of cherry crops is convincing and computationally cheap.

Genetic optimization of big data sentiment analysis

This paper deals with opinion mining from unstructured textual documents. The proposed method focuses on approach with minimum preliminary requirements about the knowledge of the analysed language and thus it can be deployed to any language. The proposed method builds on artificial intelligence, which consists of Support Vector Machines classifier, Big Data analysis and genetic algorithm optimization. To make the optimization feasible together with big data approach we have proposed GA operators, which significantly accelerate conversion to the accurate solutions. In this work we outperformed the traditional approaches (which use language dependent text preprocessing) for text valence classification with the highest achieved accuracy 90.09 %. The data set for validation was Czech texts.

Automated segmentation of powdery mildew disease from cherry leaves using image processing

An automated detection of plant disease is an important task to find features or abnormalities in plant and its effect on the fruits. In this paper an algorithm is proposed for detection of powdery mildew disease from a cherry leaf images. The proposed method uses an automated strategic removal of background from the image and then extracting the desired diseased portion. A combination of morphological operators and intensity based thresholding are used which creates a method computationally efficient and less complex. A set of public arXiv e-prints data are used to test the proposed algorithm. The tested algorithm achieves accuracy of 99%.

An efficient imaging technique for automated macula localization from fundus images

Localization of macula from fundus image plays an important role to design an automated screening tool for detection of retinal diseases. The similar color and texture of red lesions act as a bottleneck in accurate localization of macula in the fundus image. This paper presents a computer vision algorithm for automated and efficient localization of macula from low contrast and diabetic retinopathy affected fundus images. A statistical based model is used to detect macula in a specified region of fundus image which is designed using the geometric features of optic disc. The performance of the proposed algorithm of macula detection was tested on 200 normal/affected fundus images and results are significant. The computational efficiency and accurate localization of macula makes the proposed method competent enough to be used as a part of an automated screening tool for detection of retinal diseases.

Grading of colorectal cancer using histology images

This paper proposed an automated system for grading of colorectal cancer using image processing method. Almost, half a million people die every year due to colon cancer. Histopathological tissue analysis is a common method for its detection, which needs an expert pathologist. Screening for this cancer is effective for prevention as well as early detection. The method proposed segment the glands automatically by using intensity based thresholding and organizational properties for classification. In existing literature, the majority of studies based on gland segmentation in healthy or benign samples, but rarely on intermediate or high grade cancer. Unlike most of the existing methods this system is fully automated and grades the images as benign healthy, benign adenomatous, moderately differentiated malignant and poorly differentiated malignant. The proposed method achieves overall accuracy of 81% when tested on 165 histology images.

Fast localization of the optic disc in fundus images using region-based segmentation

Optic disc (OD) is considered as one of the primary features in retinal fundus images. Detection of the OD is important for the identification and severity assessment of various ophthalmic pathologies. OD localization is the first step towards accurate OD segmentation process. In this paper an adaptive region-based image segmentation method is presented for automated localization of the OD. The proposed method is tested on two publically available datasets of MESSIDOR and DRIVE. For these two datasets, the OD was successfully localized in 90 images out of 100 (90% success) and 38 images out of 40 (95 % success), respectively, with computation time of approximately 1.6 seconds per image. Experimental results indicate that the proposed method is successful in fast and robust OD localization and therefore this method can be useful for real-time automated ophthalmic pathology detection systems.

Automated segmentation of colon gland using histology images

This paper represents an automated methodology for segmentation of colon glands using histology images. The manifestations of colorectal cancer under microscope has always been challenging as staining and sectioning leads to variation in tissue specimen, which causes conflict in gland appearance. Gland segmentation and classification is very important for the automation of the system. The presented methodology automatically segments the colon gland tissues by using intensity based thresholding which makes this methodology efficient. Unlike other segmentation methods, this methodology is entirely automated and quantifies lumen and epithelial cells only in the region of interest, which makes this method computationally efficient. This methodology is efficient for calculation of number of glands as well as for segmentation of gland area and achieves overall 93.76% accuracy for both.