With the widespread use of deep-learning models in production environments, the value of deep-learning models has become more prominent. The key issues are the rights of the model trainers and the security of the specific scenarios using the models. In the commercial domain, consumers pay different fees and have access to different levels of services. Therefore, dividing the model into several shadow models with multiple weights is necessary. When holders want to use the model, they can recover the model whose performance corresponds to the number and weights of the collected shadow models so that access to the model can be controlled progressively, i.e., progressive recovery is significant. This paper proposes a neural network model secret sharing scheme (NNSS) with multiple weights for progressive recovery. The scheme uses Shamir’s polynomial to control model parameters’ sharing and embedding phase, which in turn enables hierarchical performance control in the secret model recovery phase. First, the important model parameters are extracted. Then, effective shadow parameters are assigned based on the holders’ weights in the sharing phase, and t shadow models are generated. The holders can obtain a sufficient number of shadow parameters for recovering the secret parameters with a certain probability during the recovery phase. As the number of shadow models obtained increases, the probability becomes larger, while the performance of the extracted models is related to the participants’ weights in the recovery phase. The probability is proportional to the number and weights of the shadow models obtained in the recovery phase, and the probability of the successful recovery of the shadow parameters is 1 when all t shadow models are obtained, i.e., the performance of the reconstruction model can reach the performance of the secret model. A series of experiments conducted on VGG19 verify the effectiveness of the scheme.

In recent years, the amount of data has increased explosively, which has spawned the large-scale development of cloud storage. Increasingly, individuals and enterprises store images in cloud space. The storage security of the cloud is generally guaranteed by encryption, but this can no longer meet the needs of image management and protection. In order to realize the management and loss tolerance of images, this paper proposes a thumbnail secret image sharing method. The proposed scheme combines the advantages of thumbnail-preserving encryption (TPE) and secret image sharing (SIS) with different meaningful shadows. Thumbnails can realize the visual management of stored images, and secret image sharing can realize the perfect security of stored images. The proposed scheme realizes the confidentiality, integrity, and availability of images, which are three elements of information security. Compared with TPE, our scheme not only realizes the visual management of images but also achieves loss tolerance and perfect security. Compared with SIS with different meaningful shadows, our scheme will greatly improve the sharing efficiency and reduce the consumption of computing resources. In this paper, the theoretical analysis and security proof of the proposed scheme are presented. In addition, we also conduct sufficient experiments and comparative explanations.