Abstract

Abstract

The current target tracking and detection algorithms often have mistakes and omissions when the target is occluded or small. To overcome the defects, this paper integrates bi-directional feature pyramid network (BiFPN) into cascade region-based convolutional neural network (R-CNN) for live object tracking and detection. Specifically, the BiFPN structure was utilized to connect between scales and fuse weighted features more efficiently, thereby enhancing the network’s feature extraction ability, and improving the detection effect on occluded and small targets. The proposed method, i.e., Cascade R-CNN fused with BiFPN, was compared with target detection algorithms like Cascade R-CNN and single shot detection (SSD) on a video frame dataset of wild animals. Our method achieved a mean average precision (mAP) of 91%, higher than that of SSD and Cascade R-CNN. Besides, it only took 0.42s for our method to detect each image, i.e., the real-time detection was realized. Experimental results prove that the proposed live object tracking and detection model, i.e., Cascade R-CNN fused with BiFPN, can adapt well to the complex detection environment, and achieve an excellent detection effect.

Spillover effect can lead to the free-riding behavior when joint investment takes place in the supply chain. This study examined the investment strategies of two competitive retailers who considered whether to invest a shared contract manufacturer (CM) or not. The supply chain members' operational decisions in four scenarios were analyzed through a Cournot competition model, and the paths of the retailers' investment strategies were examined. The CM's capacity portfolio optimization was NP-hard in nature, and was modelled by an investment portfolio problem. Results show that both retailers jointly invest the CM only when the difference of production costs is not high, and the intentions of joint investment will decrease when the coefficient of spillover and the degree of substitutability between products increase. The CM always benefits as long as one retailer invests, and allocates more investment on the capacity with highest revenue when he emphasizes more on the profit. For optimizing the CM's capacity portfolio problem, an artificial fish swam algorithm with uniform mutation (AFSA_UM) is developed and it shows better convergent performance and higher robustness than the basic AFSA.

No abstract