People around the world are seriously affected by the spread of the coronavirus SARS Cov2. Testing for the presence of coronavirus in the human body is based primarily on the reverse transcription polymerase chain reaction (RTPCR) recommended by the World Health Organization. Although the results of the RTPCR test are accurate, the time required for the test is not reasonable due to the high growth rate of the corona. To speed up the virus detection process, an alternative method based on image processing using CT scans of the chest is proposed in this project. Breast images are taken from the benchmark database and tested. Since this is a binary classification problem, CT scan images are applied to a deep learning model. In particular, we use a convolutional neural network (CNN) to determine if an image is "Covid-positive" or "Covid-is-negative". In addition, the benefits of Support Vector Machines (SVMs), combined with CNNs, maximize overall detection capabilities. Training and validation performance is evaluated against a database consisting of 80% training datasets and 20% validation datasets, using a total of 736 images. By changing the image properties such as height shift, width shift, zoom level, brightness level, orientation, shear, and horizontal flip, the image is initially expanded to a larger database. This brings the total to 5888 frames, including the actual 736 raw frames. Covid detection is based on a new random image that has not been used for training and validation. Accuracy, accuracy, and recognition characteristics were determined by ensemble voting of results obtained from CNNs and SVMs. CNN has achieved and training, validation, and detection accuracy, respectively. SVM provided training, testing, and detection accuracy. Combining the capabilities of deep neural networks with traditional SVMs yields a detection accuracy. The implementation of Covid prediction is done using a Python script that leverages Keras on the TensorFlow platform