In our project, the use of K-Nearest Neighbor (KNN) algorithm with transfer learning for the image classification and detection problems posed within the context of X-ray baggage security images has been considered. KNN approach requires large amounts of data to facilitate a complex end-to-end feature extraction and classification process. In the context of X-ray security screening, availability of quality dataset is a limitation. To overcome this limitation, we can employ a transfer learning paradigm such as a pre-trained KNN, primarily trained for generalized image classification tasks, where sufficient optimized training data are available can be customized for our application domain. Gabor wavelet would be used to extract the features and then these features will be used to train different images in the dataset using KNN. We would empirically show that fine-tuned KNN features yield superior performance when compared with the standard object classification tasks.