Abstract: One of the most important points to improve abundance estimation for linear mixture spectral model lies in end-member spectral constituent. A novel approach to improve abundance estimation of hyper-spectral image using spectral piecewise constant features is presented. This method firstly extracts the spectral features by partitioning the spectral signals into a fixed number of contiguous intervals with constant intensities in terms of minimizing the mean square error. Then, the estimation is performed by unmixing the pixel in the feature space with constrained least square algorithm to achieve the respective abundance fractions of these end-members present in the pixel. Algorithm validation and comparison were done with simulated and real data. Experimental results demonstrate the proposed method can significantly improve the least squares estimation of end-member abundances using remotely sensed hyper-spectral signals, as compared to those of original hyper-spectral signals or discrete wavelet transform based features.