A New Application of Pixel Purity Index to Unsupervised Multispectral Image Classification
نویسندگان
چکیده
Two major challenging issues arise in unsupervised classification. One is how to generate desired knowledge directly from the data in an unsupervised manner. The other is how to find an appropriate follow-up classifier to use the obtained unsupervised knowledge to perform supervised classification. This paper presents a new approach to unsupervised classification for multispectral imagery. To address the first issue the pixel purity index (PPI) which is commonly used in hyperspeftral imaging for endmember extraction is used to find a good set of initial training samples without prior knowledge. To address the second issue the PPI-found samples are then used as training samples for a support vector machine to find a good set of training samples for a follow-up supervised classifier, Fisher’s linear discriminate analysis (FLDA) which performs classification iteratively to produce final results. The experimental results show the proposed approach has great promise in unsupervised classification.
منابع مشابه
Recognation and separation of high purity calcite mineral regions in carbonate units using Aster multispectral data and Sentinel 2 (case study northwest of Shahrekord)
Remote sensing has found a special technique and effective in the geological studies and mineral identification for the determination of minerals in the primary detection. The purpose of this study is to recognize the high purity calcite mineral regions in carbonate units using Aster and sentinel 2 images. Sampling of rocks and laboratory analysis using XRF and XRD for verification. The results...
متن کاملSelf-organizing Map for Clustering of Remote Sensing Imagery
We present a neural unsupervised pattern recognition approach for two applications related to significant topics of Earth Observation (EO) imagery: (a) EO image region classification; (b) multispectral pixel classification. The proposed model is based on the Self-Organizing Map (SOM) clustering, which is compared to two benchmark unsupervised classifiers: k-means and fuzzy c-means. We propose t...
متن کاملCombining of Magnitude and Direction of Change Indices to Unsupervised Change Detection in Multitemporal Multispectral Remote Sensing Images
In remote sensing, image-based change detection techniques, analyze two images acquired over the same area at different times t1 and t2 to identify the changes occurred on the Earth's surface. Change detection approaches are mainly categorized as supervised and unsupervised. Generating the change index is a key step for change detection in multi-temporal remote sensing images. Unsupervised chan...
متن کاملAn unsupervised marker image generation method for watershed segmen- tation of multispectral imagery
Watershed transformation in mathematical morphology is a powerful tool for image segmentation. Watershed transformation based segmentation is generally a marker-controlled segmentation. This paper proposes a novel method of maker image generation based on unsupervised ISODATA classification. The method incorporates spectral information contained in multispectral data into watershed transformati...
متن کاملRemotely Sensed LANDSAT Image Classification Using Neural Network Approaches
In paper, LANDSAT multispectral image is classified using several unsupervised and supervised techniques. Pixel-by-pixel classification approaches proved to be infeasible as well as time consuming in case of multispectral images. To overcome this, instead of classifying each pixel, feature based classification approach is used. Three supervised techniques namely, k-NN, BPNN and PCNN are investi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010