• K. AMARAVATHI
• N. LOKESWARI

Now a days for identifying or predict any living beings, we should have proper idea about that living beings. For example if we want to predict some animals,plants,birds,dogs and so on, we need to know clearly about that living beings, before we try to predict the end result. In general we try to take images of those living beings especially birds and then try to classify the birds based on its features and then try to decide the species name. In this proposed application we develop a fine-grained image classifier using a general deep convolutional neural network (D-CNN). In order to increase the accuracy of our proposed model we try to modify the DCNN model from the following two aspects. First, to better model the h-level hierarchical label structure of the fine-grained image classes contained in the given training data set, we introduce h fully connected (fc) layers to replace the top fc layer of a given DCNN model and train them with the cascaded softmax loss. Second, we propose a novel loss function, namely, generalized large-margin (GLM) loss, to make the given DCNN model explicitly explore the hierarchical label structure and the similarity regularities of the fine-grained image classes. The GLM loss explicitly not only reduces between-class similarity and within-class variance of the learned features by DCNN models but also makes the subclasses belonging to the same coarse class be more similar to each other than those belonging to different coarse classes in the feature space. By implementing these two aspects on current DCNN model, we try to propose a new modified DCCN model known as MDCNN model, which can increase the accuracy of deep learning models in order to predict the bird species very accurately. By conducting various experiments on our proposed model by comparing with several pre-trained models such as AlexNet, GoogLeNet, and VGG, using three benchmark data sets.Our simulation results clearly demonstrate the accuracy and efficiency of our modified MDCNN model is standing on top of all primitive models.

Deep Convolutional Neural Network, Cascaded Softmax Loss, Generalized Large-Margin (GLM) Loss, Fully Connected.

"Bird Species Identification Using Modified Deep Learning (CNN) Model by Integrating Cascaded Softmax Layer", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 5, page no.g793-g802, May-2021, Available :http://www.jetir.org/papers/JETIR2105903.pdf

"Bird Species Identification Using Modified Deep Learning (CNN) Model by Integrating Cascaded Softmax Layer", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 5, page no. ppg793-g802, May-2021, Available at : http://www.jetir.org/papers/JETIR2105903.pdf