The objective of this study is to employ deep learning techniques to aid radiologists in improving the efficiency and accuracy of breast cancer detection. This paper involves a deep neural network (DNN) classifier method in order to classify breast cancer in the MIAS dataset. Initially, we begin by preprocessing the mammography pictures to eliminate the digitization noise using a Wiener filter and enhance the contrast of the breast tissue using the contrast stretching approach. Subsequently, the Otsu’s method is employed to detect the Region of Interest (ROI), from which 24 features are retrieved using the GLCM technique and color models. The subsequent step involves training a deep neural network (DNN) classifier utilizing the important features from the training set. The trained DNN classifier is next employed to classify the Regions of Interests (ROIs) in the test set. We utilized the mini-MAIS dataset for experimental purposes. Our research shows that the suggested DNN classifier achieves an average accuracy of 93.24% in classifications. Significant values were also found for the remaining output parameters, which include sensitivity, accuracy, recall, F-measure. With a score of 0.9324 for accuracy, 0.9315 for specificity, 0.8748 for precision, 0.8480 for recall, and an F-measure of 0.8547, the system was more effective than other classifiers. After analyzing these measures, it is clear that the proposed DNN classifier performs better than current state-of-the-art approaches.