The security of Software Defined Networking (SDN) is becoming increasingly critical as it develops. The centralized management of SDN allows attackers to launch a variety of intrusions that will consume controller and switch memory, network bandwidth, and server resources. This paper presents a new way to recognize an intrusion scenario using a Deep Learning (DL) method based on sequential image processing. In this method, we propose to present an intrusion scenario through a sequence of sequential images. Since the intrusion scenario is a sequence of malicious activities, we propose to represent each activity by an image, and the intrusion scenario is represented by a sequence of images that are related to each other. We propose to use a hybrid deep learning method based on Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) to predict the image sequences related to the intrusion scenario. To demonstrate the efficacy of the suggested approach, we apply it to a case study and compare it with single image classification. The experimental results demonstrate that the suggested approach can accurately identify an intrusion scenario.