To improve the performance of the detection of prostate cancer, multi-parametric(mp) MRI is increasingly mentioned due to its possibility of combining anatomical information with functional information. In this paper, we propose an automatic prostate cancer detection method using signal intensity, intensity histogram and texture features based on SVM(Support Vector Machine) in multi-parametric MR images. For each T2wMR slice, a ROI for a prostate cancer was marked by a radiologist referring to the result of co-registration of T2wMR and histopathology image. In the automatic detection of prostate cancers, first, to consider the corresponding prostate cancer ROI on DWI and ADC map to T2wMR, the transformation parameters of DWI are estimated by normalized mutual information-based rigid registration and the estimated transformations were applied to ADC map. Then, to normalize the signal intensity range among inter-patient images, histogram stretching is performed. Second, to extract consistent intensity and texture features among inter-patient multi-parametric images due to their various signal intensity range, intensity histogram normalized the range of signal intensity among inter-patient images via histogram stretching. And from the prostate cancer ROIs of training set, intensity, gray-level co-occurrence features were extracted. to differentiate hemorrhage areas from prostate cancer on T2wMR, histogram and texture features were considered on T1wMR because hemorrhage areas on T1wMR have high signal intensities. Third, to detect prostate cancer areas in T2wMR, SVM classifier with selected feature was trained by selected features on mpMRI. For evaluation of our automatic prostate cancer detection method, leave-one-out cross validation was performed and our results were evaluated by accuracy, sensitivity, specificity and DSC. When considering T2wMR and ADC map together were improved when T2wMR alone, And When considering T2wMR, ADC map and T1wMR features together were improved T2wMR and ADC map features. Our feature classification using mpMRI can improve the prostate detection rate of T2wMR. Our method can be used to improve the performance of detection and staging of prostate cancer on mpMRI.