Cloud computing has become an integral part of our lives, allowing us to store and access data quickly and easily. Unfortunately, cybercriminals have developed increasingly sophisticated techniques to exploit network vulnerabilities and bring them down. This paper proposes using two pre-trained deep learning models, VGG16 and ResNet50, to classify benign traffic and DrDoS (distributed denial-of-service) attacks in cloud environments. The Cloud Attacks Dataset, sourced from the CICIDS 2019 dataset, includes a comprehensive collection of network traffic instances represented in binary visualization format and is divided into two classes: benign traffic and DrDoS attacks. The dataset has been split into a training set of 5000 images and a testing set of 1168 images with equal representation of benign traffic and DrDoS attack instances. We aim to provide a robust and reliable solution to detect DrDoS attacks and protect networks from malicious activities.