This paper proposes a method for the classification of urban hyperspectral data using Support Vector Machines (SVM) and morphological profiles. The approach combines both spatial and spectral information to improve classification accuracy. The method involves the fusion of morphological information and original hyperspectral data, where the two feature vectors are concatenated into one. After dimensionality reduction, the final classification is performed using an SVM classifier. The method is tested on ROSIS data from urban areas and shows significant improvements in classification accuracy compared to previous approaches based on morphological profiles and conventional spectral classification. For example, overall accuracy increased from 79% to 83% without feature reduction and to 87% with feature reduction. The method also performs well with a limited training set. The approach uses feature extraction techniques such as Decision Boundary Feature Extraction (DBFE) and Nonparametric Weighted Feature Extraction (NWFE) to reduce redundancy and improve classification performance. The method is compared to statistical classification methods and SVM classification, and experimental results on two high-resolution remote sensing data sets from urban areas demonstrate its effectiveness. The proposed method outperforms previous approaches in terms of classification accuracy and is suitable for applications requiring high spatial resolution and accurate classification of urban areas. The method is based on the fusion of spatial and spectral information, and it uses SVM for classification. The results show that the method is effective in classifying urban areas with high accuracy and is suitable for applications requiring high spatial resolution and accurate classification of urban areas.This paper proposes a method for the classification of urban hyperspectral data using Support Vector Machines (SVM) and morphological profiles. The approach combines both spatial and spectral information to improve classification accuracy. The method involves the fusion of morphological information and original hyperspectral data, where the two feature vectors are concatenated into one. After dimensionality reduction, the final classification is performed using an SVM classifier. The method is tested on ROSIS data from urban areas and shows significant improvements in classification accuracy compared to previous approaches based on morphological profiles and conventional spectral classification. For example, overall accuracy increased from 79% to 83% without feature reduction and to 87% with feature reduction. The method also performs well with a limited training set. The approach uses feature extraction techniques such as Decision Boundary Feature Extraction (DBFE) and Nonparametric Weighted Feature Extraction (NWFE) to reduce redundancy and improve classification performance. The method is compared to statistical classification methods and SVM classification, and experimental results on two high-resolution remote sensing data sets from urban areas demonstrate its effectiveness. The proposed method outperforms previous approaches in terms of classification accuracy and is suitable for applications requiring high spatial resolution and accurate classification of urban areas. The method is based on the fusion of spatial and spectral information, and it uses SVM for classification. The results show that the method is effective in classifying urban areas with high accuracy and is suitable for applications requiring high spatial resolution and accurate classification of urban areas.