scholarly journals Analisis Perbandingan KNN dengan SVM untuk Klasifikasi Penyakit Diabetes Retinopati berdasarkan Citra Eksudat dan Mikroaneurisma

Author(s):  
SUCI AULIA ◽  
SUGONDO HADIYOSO ◽  
DADAN NUR RAMADAN

ABSTRAKPenelitian mengenai pengklasifikasian tingkat keparahan penyakit Diabetes Retinopati berbasis image processing masih hangat dibicarakan, citra yang biasa digunakan untuk mendeteksi jenis penyakit ini adalah citra optik disk, mikroaneurisma, eksudat, dan hemorrhages yang berasal dari citra fundus. Pada penelitian ini telah dilakukan perbandingan algoritma SVM dengan KNN untuk klasifikasi penyakit diabetes retinopati (mild, moderate, severe) berdasarkan citra eksudat dan microaneurisma. Untuk proses ekstraksi ciri digunakan metode wavelet  pada masing-masing kedua metode tersebut. Pada penelitian ini digunakan 160 data uji, masing-masing 40 citra untuk kelas normal, kelas mild, kelas moderate, kelas saviere. Tingkat akurasi yang diperoleh dengan menggunakan metode KNN lebih tinggi dibandingkan SVM, yaitu 65 % dan 62%. Klasifikasi dengan algoritma KNN diperoleh hasil terbaik dengan parameter K=9 cityblock. Sedangkan klasifikasi dengan metode SVM diperoleh hasil terbaik dengan parameter One Agains All.Kata kunci: Diabetic Retinopathy, KNN , SVM, Wavelet. ABSTRACT Research based on severity classification of the disease diabetic retinopathy by using image processing method is still hotly debated, the image is used to detect the type of this disease is an optical image of the disk, microaneurysm, exudates, and bleeding of the image of the fundus. This study was performed to compare SVM method with KNN method for classification of diabetic retinopathy disease (mild, moderate, severe) based on exudate and microaneurysm image. For feature extraction uses wavelet method, and each of the two methods. This study made use of 160 test data, each of 40 images for normal class, mild class, moderate class, severe class. The accuracy obtained by KNN higher than SVM, with 65% and 62%. KNN classification method achieved the best results with the parameters K = 9, cityblock. While the classification with SVM method obtained the best results with parameters One agains all .Keywords: Diabetic Retinopathy, KNN, SVM, Wavelet.

2017 ◽  
Author(s):  
Maziyar M. Khansari ◽  
William O’Neill ◽  
Richard Penn ◽  
Norman P. Blair ◽  
Felix Chau ◽  
...  

2019 ◽  
Vol 9 (7) ◽  
pp. 1385 ◽  
Author(s):  
Luca Donati ◽  
Eleonora Iotti ◽  
Giulio Mordonini ◽  
Andrea Prati

Visual classification of commercial products is a branch of the wider fields of object detection and feature extraction in computer vision, and, in particular, it is an important step in the creative workflow in fashion industries. Automatically classifying garment features makes both designers and data experts aware of their overall production, which is fundamental in order to organize marketing campaigns, avoid duplicates, categorize apparel products for e-commerce purposes, and so on. There are many different techniques for visual classification, ranging from standard image processing to machine learning approaches: this work, made by using and testing the aforementioned approaches in collaboration with Adidas AG™, describes a real-world study aimed at automatically recognizing and classifying logos, stripes, colors, and other features of clothing, solely from final rendering images of their products. Specifically, both deep learning and image processing techniques, such as template matching, were used. The result is a novel system for image recognition and feature extraction that has a high classification accuracy and which is reliable and robust enough to be used by a company like Adidas. This paper shows the main problems and proposed solutions in the development of this system, and the experimental results on the Adidas AG™ dataset.


2015 ◽  
Vol 9 (7) ◽  
pp. 783-789 ◽  
Author(s):  
Yang Woo Yong ◽  
Park Ji Hoon ◽  
Bae Jun Woo ◽  
Kang Sung Cheol ◽  
Myung Noh Hoon

Author(s):  
Shaziya Banu S ◽  
Ravindra S

<p>Diabetic Retinopathy (DR) is a related malady with diabetes and primary driver of sightlessness in diabetic patients. Epidemiological overview categorizes DR among four significant reasons for sight impedance. DR is a microvascular entanglement in which meager retinal veins may blast, bringing about vision misfortune. In this condition veins in retina swells and may blast in severe extreme condition. Operative medication is timely discovery by steady screenings that is by emphasizing the determination of retinal images using appropriate image processing techniques such as, Preprocessing of retinal image, image segmentation using sobel edge detector, local features extraction like mean, standard deviation, variance, Entropy, histogram values and so on. For classification of retina, system uses K-Nearest Neighbor (KNN) classifier. By adopting this approach, The classification of normal and abnormal images of retina is easy and will reduce the number of reviews for the ophthalmologists. Developing a method to automate functionality of retinal examination helps doctor to identify patient’s condition on disease. So that they can medicate the disease accordingly.</p>


2018 ◽  
Vol 1 (2) ◽  
pp. 46
Author(s):  
Tri Septianto ◽  
Endang Setyati ◽  
Joan Santoso

A higher level of image processing usually contains some kind of classification or recognition. Digit classification is an important subfield in handwritten recognition. Handwritten digits are characterized by large variations so template matching, in general, is inefficient and low in accuracy. In this paper, we propose the classification of the digit of the year of a relic inscription in the Kingdom of Majapahit using Support Vector Machine (SVM). This method is able to cope with very large feature dimensions and without reducing existing features extraction. While the method used for feature extraction using the Gray-Level Co-Occurrence Matrix (GLCM), special for texture analysis. This experiment is divided into 10 classification class, namely: class 1, 2, 3, 4, 5, 6, 7, 8, 9, and class 0. Each class is tested with 10 data so that the whole data testing are 100 data number year. The use of GLCM and SVM methods have obtained an average of classification results about 77 %.


Sign in / Sign up

Export Citation Format

Share Document