Blockchain technology (BCT) provides a new way to mitigate the default risks of lease contracts resulting from the information asymmetry in leasing. The conceptual architecture of a consortium blockchain-based leasing platform (CBLP) is first proposed to facilitate information sharing between small and medium-sized enterprises (SMEs, the “lessees”) and leasing firms (LFs, the “lessors”). Then, based on evolutionary game theory (EGT), this study builds a two-party game model and analyzes the influences of four types of factors (i.e., information sharing, credit, incentive–penalty, and risk) on SMEs’ contract compliance or default behaviors with/without blockchain empowerment. The primary findings of this study are as follows: (1) SMEs and LFs eventually evolve to implement the ideal “win–win” strategies of complying with the contract and adopting BCT. (2) The large residual value of the leased asset can tempt SMEs to conduct a default action of unauthorized asset disposal, while leading LFs to access the CBLP to utilize information shared on-chain. (3) When the maintenance service is outsourced instead of being provided by lessors, the maintenance fee is not a core determinant affecting the equilibrium state. (4) There is a critical value concerning the default penalty on-chain to incentivize the involved parties to keep their commitments. (5) The capability of utilizing information, storage overhead, and security risk should all be taken into consideration when deciding on the optimal strategies for SMEs and LFs. This study provides comprehensive insights for designing an incentive mechanism to encourage lessees and lessors to cooperatively construct a sustainable and trustworthy leasing environment.

Effectively utilizing the common information in a set of video frames is a vital aspect in video segmentation. However, existing methods which transport the common information from a prior frame to the current frame do not make use of the common information effectively. In order to address this issue, we apply a new strategy that jointly segments object through convolutional neural network (CNN) to build a proposal-driven framework for exploiting the common information between two video frames by processing two video frames simultaneously in this letter. Moreover, proposals from the video frames are found useful for refining the segmentation results though fusing their segmentation results with the ones of the video frames. In our framework, proposals with features are generated by Faster R-CNN and the L2 loss function is used to establish proposal pairs among proposals from the two selected frames. A new trained ResNet then keeps proposal pairs which contain the same content and the PSPNet model for segmentation is utilized to generate the segmentation results belonging to the frames and proposals. Finally, the proposals' segmentation results are refined using the video frames' segmentation results. The VOT 2016 segmentation dataset, DAVIS 2017 dataset, and SegTrack v2 dataset were used for training and testing our framework. Experimental results show that our proposal-driven segmentation framework is able to achieve higher accuracies in video segmentation challenge compared to those of the existing video segmentation methods.

In recent years, related research has focused on how to safely transfer and protect the privacy of images in social network services while providing easy access by authorized users. To safeguard privacy, we suggest an image encryption scheme that combines data hiding and image encryption. The proposed scheme successfully decrypts images after JPEG compression attacks and preserves the privacy of secret regions through the use of block scrambling encryption based on region selection. Simultaneously, the scheme can handle nonuniform secret regions and obtain more sensitive secret keys because of the incorporation of a chaotic system. The enhanced deep learning-based data-hiding technology reduces algorithm complexity by enabling the encryption position to be determined in the decryption phase without the need for any information or equipment. However, this approach also increases algorithm security, because only when the right secret data are extracted can they be decrypted successfully. According to the experimental findings, the proposed scheme can correctly decrypt images via JPEG compression while maintaining visually acceptable quality. The proposed scheme can achieve greater robustness against image processing algorithms and a wider secret key space than traditional schemes.

In recently years, some useful unsupervised video object segmentation methods that emphasize the common information in videos have been proposed. Despite the effectiveness of these methods, they ignore the information from the shallow layers of the network and thus fail to segment the details of the objects. To address this problem, we propose a multi-attention network for unsupervised video object segmentation (MANet). Recent studies show that the deep layers of networks are sensitive to high-level semantic information but messy details, while it is opposite for shallow layers. From this insight, a multi-attention module is designed by taking into account the information from the shallow layers in addition to that from the deep layers. This module can distinguish the primary object and segment the details of the object effectively by enhancing the common information between video frames while combing the features from the shallow layers and the deep layers. Experimental results on the DAVIS-2016 and SegTrack v2 datasets show that our network outperforms the state-of-the-art methods.