In the era of deep learning as a service, ensuring that model services are sustainable is a key challenge. To achieve sustainability, the model services, including but not limited to storage and inference, must maintain model security while preserving system efficiency, and be applicable to all deep models. To address these issues, we propose a sub-network-based model storage and inference solution that integrates blockchain and IPFS, which includes a highly distributed storage method, a tamper-proof checking method, a double-attribute-based permission management method, and an automatic inference method. We also design a smart contract to deploy these methods in the blockchain. The storage method divides a deep model into intra-sub-network and inter-sub-network information. Sub-network files are stored in the IPFS, while their records in the blockchain are designed as a chained structure based on their encrypted address. Connections between sub-networks are represented as attributes of their records. This method enhances model security and improves storage and computational efficiency of the blockchain. The tamper-proof checking method is designed based on the chained structure of sub-network records and includes on-chain checking and IPFS-based checking stages. It efficiently and dynamically monitors model correctness. The permission management method restricts user permission based on the user role and the expiration time, further reducing the risk of model attacks and controlling system efficiency. The automatic inference method is designed based on the idea of preceding sub-network encrypted address lookup. It can distribute trusted off-chain computing resources to perform sub-network inference and use the IPFS to store model inputs and sub-network outputs, further alleviating the on-chain storage burden and computational load. This solution is not restricted to model architectures and division methods, or sub-network recording orders, making it highly applicable. In experiments and analyses, we present a use case in intelligent transportation and analyze the security, applicability, and system efficiency of the proposed solution, particularly focusing on the on-chain efficiency. The experimental results indicate that the proposed solution can balance security and system efficiency by controlling the number of sub-networks, thus it is a step towards sustainable model services for deep learning.

Model evaluation is critical in deep learning. However, the traditional model evaluation approach is susceptible to issues of untrustworthiness, including insecure data and model sharing, insecure model training, incorrect model evaluation, centralized model evaluation, and evaluation results that can be tampered easily. To minimize these untrustworthiness issues, this paper proposes a blockchain-based model evaluation framework. The framework consists of an access control layer, a storage layer, a model training layer, and a model evaluation layer. The access control layer facilitates secure resource sharing. To achieve fine-grained and flexible access control, an attribute-based access control model combining the idea of a role-based access control model is adopted. A smart contract is designed to manage the access control policies stored in the blockchain ledger. The storage layer ensures efficient and secure storage of resources. Resource files are stored in the IPFS, with the encrypted results of their index addresses recorded in the blockchain ledger. Another smart contract is designed to achieve decentralized and efficient management of resource records. The model training layer performs training on users’ servers, and, to ensure security, the training data must have records in the blockchain. The model evaluation layer utilizes the recorded data to evaluate the recorded models. A method in the smart contract of the storage layer is designed to enable evaluation, with scores automatically uploaded as a resource attribute. The proposed framework is applied to deep learning-based motion object segmentation, demonstrating its key functionalities. Furthermore, we validated the storage strategy adopted by the framework, and the trustworthiness of the framework is also analyzed.

Ship detection in synthetic aperture radar (SAR) images faces challenges such as strong background interference, varying ship appearance, and distribution and high real-time requirements. Although attention-based deep learning methods dominate this field, the design of lightweight models with efficient attention mechanisms capable of addressing the aforementioned challenges remains underexplored. To address this issue, we propose a lightweight SAR ship detection model named LSDFormer, which is built upon the MetaFormer architecture and consists of an efficient multiattention-enhanced backbone and neck and a structural reparameterization (SR)-enhanced head. We employ two lightweight modules for the backbone and neck: a PoolFormer-based feature extraction module with efficient channel modulation attention is proposed to enhance ship features and suppress background interference, and a downsampling module using efficient channel aggregation attention and group convolutions is introduced to enrich ship features. The position-sensitive attention from YOLOv11 is also introduced to handle variations in ship appearance and distribution. These three attentions are integrated into an efficient multiattention mechanism. Furthermore, an SR-based detection branch is proposed for the head of LSDFormer, which enhances ship features while reducing model complexity. Extensive experiments on SSDD and HRSID datasets demonstrate the superiority and effectiveness of LSDFormer, achieving AP50 of $\mathbf {98.5\pm 0.4\%}$ and $\mathbf {92.8\pm 0.2\%}$, respectively, with only $\mathbf {1.5}$ M parameters and $\mathbf {4.1}$ GFLOPs. The average processing time per image is $\mathbf {4.9}$ ms on SSDD and $\mathbf {4.2}$ ms on HRSID, confirming its real-time performance.