Breast common cancer kind of cancer that affects women globally. Around 30% of all new cases of cancer in women are anticipated to be breast cancer by 2022. This life-threatening disease is an incurable disease but controllable. Nevertheless, early diagnosis through routine inspection can boost recovery and survival chances. A computer-aided breast cancer diagnosis can show promising results to automatically classify breast histopathology images. The classification of benign and malignant patients using mammography pictures is suggested in this work utilizing a mix of curvelet transformations and support vector machines. Curvelet transformation fetches the optimal features for breast cancer classification. Images of Mammograms used in this paper for evaluation is provided by two databases i.e., Break His and BisQUE. The model’s performance is measured in form of average and standard deviation of accuracy, False Negative Rate, True Positive Rate, and AUC of ROC. Results show the remarkable performance of the model by achieving accuracy of 91.0 %, and 85.9%, for Break His and BisQUE respectively.