Malware refers to a wide range of malicious programs created to damage the software, hardware, and network components, or compromise data in computers and other devices. It contains viruses, worms, Trojans, ransomware, spyware, and adware. In this research, it has been possible to propose an effective strategy for identifying and classifying malware. Smaller sample sizes and multi-label classification were avoided; instead; the use of the convolutional neural network (CNN) and the ResNet architectural approach was used to identify and classify the malware sample. The steps aimed at converting the malicious files into byte files and then into the image that would be used further for training the deep learning models. Solving this problem required the use of image-based training to enhance the level of accuracy in the detection and identification of malware within various classes. The integration of CNNs and ResNet guarantees good feature extraction while the latter improves deep learning characteristics in the proposed IPE-DEEP model. One of the major contributions of the proposed work is to treat byte files as images to make our learning more accurate and practical for malware analysis in the real world.