Preprocessing is an essential part to achieve good segmentation since it affects the feature extraction process. Melanoma have various shapes and their extracted features from image are used for early stage detection. Due to the fact that melanoma is one of dangerous diseases, early detection is required to prevent further phase of cancer from developing. In this paper, we propose a new framework to detect cancer on skin images using color feature extraction and feature selection. The default color space of skin images is RGB, then brightness is added to distinguish the normal and darken area on the skin. After that, average filter and histogram equalization are applied as well for attaining a good color intensities which are capable of determining normal skin from suspicious one. Otsu thresholding is utilized afterwards for melanoma segmentation. There are 147 features extracted from segmented images. Those features are reduced using three types of feature selection algorithms: Linear Discriminant Analysis (LDA), Correlation based Feature Selection (CFS), and Relief. All selected features are classified using k-Nearest Neighbor  (k-NN). Relief is known to be the best feature selection method among others and the optimal k value is 7 with 10-cross validation with accuracy of 0.835 and 0.845, without and with feature selection respectively. The result indicates that the frameworks is applicable for early skin cancer detection.

Content Based Image Retrieval (CBIR) is a process to search for an image based on the content or features that are inside. Nowadays, many image retrieval applications have been made to meet the needs, so this application can provide convenience in terms of the introduction and search for an image. In this research, we used 10 different objects as image retrieval consists of Bicycle, Cow, Flower, Frangipani, Grape, Horse, Lovebird, Orange, Strawberry, Tree. These objects can be expressed in 10 classes. Our aim using these objects is viewed from the color of every object and the object of a different kind. From that point of view we built a CBIR system by utilizing the main features of the object (image). The main feature is the color feature. In this research, the main process is the extraction of color features with the color histogram and color moments. So in this research it will produce feature extraction by measuring the similarity in the library image. This measurement is done by calculating the closest distance using the euclidean distance method. The data library used in this research is 10 pieces of data, the test data is 50 pieces with 5 pieces for each of these classes. After testing using data and methods described above, the results of accuracy are obtained that the application of the color moments method gets better results than the color histogram method.

No abstract

No abstract

No abstract