Tuberculosis is one of the top 10 causes of death worldwide. In order to reduce the workload of doctors and the probability of human error, this paper presents an automatic detection algorithm for Mycobacterium tuberculosis in Ziehl-Neelsen Sputum Smear Images. The algorithm uses color feature and three morphological characters, which are aspect ratio, circularity and area. Background Equalization algorithm is proposed to utilize color feature sufficiently. This algorithm takes advantage of the watershed algorithm and the channel “a” in Lab color space. Experimental results confirmed the high accuracy of the proposed algorithm.