Flower identification and classification is a challenging job because of the extensive variety of flower species. Horticulture and textile industry uses flower classification for their applications. In this paper, we have analyzed the current growth in transfer learning using convolutional neural network. We have proposed a transfer learning model architecture that uses MobileNetV2 model. The enhanced EMobileNet model uses RMSProp and SparseCategoricalCrossentropy algorithms as optimizers and losses respectively for flower classification on a Kaggle flower dataset. The advantage of RMSProp optimizer is that it converges fast and has a stable learning. Sparse categorical cross-entropy is a loss algorithm that can be used for a number of multiclass classification problems. The classification accuracy is measured. It is found that EMobileNet performs better than CNN. This classification application can be utilized in textile industry where different patterns of flowers are required to be printed on the cloth material, which can increase the production of variety of clothes.