Segmentation of near-infrared (NIR) microscopic image by feature extraction and clustering analysis methods can be used for efficient extraction of chemical information. Due to the high computational complexity of principal component analysis (PCA) in extracting features, we propose a weighted two-dimensional PCA (W2DPCA) spectral feature extraction scheme in this paper, which is combined with fuzzy C-mean (FCM) algorithm to extract the chemical information of NIR microscopic image. The feasibility and effectiveness of the proposed algorithm are verified by simulation experiments performed on NIR microscopic image of tablets. Experimental results show that W2DPCA-FCM is an effective infrared microscopy image analysis method since it can reduce the computation time and improve the clustering accuracy.