This study proposes picture enhancement in conjunction with enhancement through a deep convolution generative adversarial network to enhance categorization of cervical cancer images. With this method (DCGAN), the [cervical cancer] images can be classified in a more precise way. In order to start the right treatment, cervical cancer, a leading cause of cancer-related death for women globally, has to be detected early. Cervical cancer may be difficult to diagnose in photos because of its complexity and wide range of symptoms. We propose to generate synthetic pictures that are comparable to each category in the current dataset using a deep convolutional generative adversarial network (DCGAN). We use three different dataset layers to train the DCGAN for each class in a different session. They are then used once again to train a deep learning model for the categorization of pictures of cervical cancer after the production of fake images. These produced pictures together with the augmented dataset are used to in the pre-training of the classification model. The next conventional practice during the transfer learning technique is to further train the classifier on the training images. The proposed method is applied using a cervical cancer images dataset, which demonstrated that the combination of transfer learning and artificial data generated through DCGAN is more accurate than using only the original dataset.