No abstract

In the automatic apple sorting task, it is necessary to automatically classify certain apple species. A shallow convolutional neural network (CNN) architecture is proposed for this purpose. After collecting a certain number of apple images and labelling them, training data is obtained through a series of data augmentation operations, and then training and parameter optimization are carried out through the Caffe framework. The feasibility of the method is verified by experiments which are divided into two cases. In the case of no occlusion, the classification accuracy of apple images reaches approximately 92% in our test set. Besides, block voting is used to aid the proposed method and a good result can be achieved in our test set in the case of part occlusion caused by branches and leaves, rotten spots, and other kinds of apples. The proposed shallow network is characterized by a small number of parameters and shows resistance to overfitting with a limited dataset. Such a network presents an alternative for classification related tasks in smart visual Internet of Things and brings attention to reducing the complexity of deep neural networks while maintaining their strength. INDEX TERMS Image classification, CNN, overfitting, smart visual Internet of Things.

No abstract

Simulation on crowd evacuation has attracted plenty of attention in the past few years. Emotion plays a significant role in human decision making processes and on the behaviors during the evacuation. Moreover, grouping is a common phenomenon in crowds, and there are a lot of roles included in a group. In this paper, we present a unified framework to model the structure aspect of the groups with emotion for a crowd suffering fires and explosions in public areas. The proposed evacuation framework consists of two components: (1) an emotion model for various roles in the case of an emergency. Emotional changing is a complex process where different roles are influenced by different factors. (2) Both intra-group structure and inter-group relationship are considered to simulate the herd phenomenon in emergency. The structure and behaviors of a group are related to the emotions of different roles. A follower can decide whether to leave the group or not according to the emotion of the group leader and the one of his own, thus the members of the group may always dynamically change. Simulation results show that the developed framework can practically model the emotion changing and different group structures of evacuation crowds.

No abstract