In order to enhance the image quality of the detector under the condition of Low Level Light(LLL) signal, a method of high sensitivity imaging based on Digital Micromirror Device(DMD) coding Hadamard Transform was proposed. On the basis of the assumptions that the noise of the detector and the signal as well as the noise measured each time are both independent, the influence of the noise to the SNR was analyzed in both HT coding imaging and classical imaging. The result was deduced that the improvement of SNR was related to the length n of coding templates in coding imaging, and the value was sqrt(n)/2 times of that in classic imaging. In order to reduce the encoding time of the image, the partitions coding method was applied. Results show that, compared with classical imaging modalities, the use of partitions coding Hadamard transform imaging significantly improves the signal to noise ratio of the image, while the coding time of high-resolution images is shortened.