Effective diagnosis of skin tumours mainly relies on the analysis of the characteristics of the lesion. Automatic detection of malignant skin lesion has become a mandatory task to reduce the risk of human deaths and increase their survival. This article proposes a study of skin lesion classification using transfer learning approach. The transfer learning model uses four different state-of-the-art architectures, namely Inception v3, Residual Networks (ResNet 50), Dense Convolutional Networks (DenseNet 201) and Inception Residual Networks (Inception ResNet v2). These models are trained under the dataset comprising seven different classes of skin lesions. The skin lesion images are pre-processed using image quantization, grayscaling and the Wiener filter before final training step. These models are compared for performance evaluation on different metrics. The present study shows the efficacy of the methodology for automated classification of lesion images.

No abstract

In this paper, we focus on one of the visual recognition facets of computer vision, i.e. image captioning. This model’s goal is to come up with captions for an image. Using deep learning techniques, image captioning aims to generate captions for an image automatically. Initially, a Convolutional Neural Network is used to detect the objects in the image (InceptionV3). Recurrent Neural Networks (RNN) and Long Short Term Memory (LSTM) with attention mechanism are used to generate a syntactically and semantically correct caption for the image based on the detected objects. In our project, we're working with a traffic sign dataset that has been captioned using the process described above. This model is extremely useful for visually impaired people who need to cross roads safely.