Text-based CAPTCHAs are friendly and easy to understand, which have been widely used in the security defense mechanism of many Internet applications. Traditional text-based CAPTCHAs improve security by distorting characters or adding background noise. With the development of deep learning, its security is threatened and over-deformed characters will bring new problems to human. To address this, this paper designs a perturbation optimization framework with randomization strategy for text-based CAPTCHAs generation (denoted as PORG), which is friendly for human but difficult for machines. Specifically, the proposed PORG devises a perturbation generation network (PGN) based on current advanced and stable perturbation methods to construct multiple perturbation factors and applies a randomization strategy to generate diverse perturbed images. In particular, the perturbation factors generated by existing methods destroy the visual information conveyed by the CAPTCHA images. To this end, a perturbation optimization network (PON) is designed to control the introduced perturbation factors by extending the distance at feature-level and narrowing the gap at global-level, which makes the generated CAPTCHAs remain human-friendly while effectively treating the attacker model. Extensive experiments conducted on eight real-world datasets show the outperformance of the proposed PORG (e.g., attack accuracy is dropped from 90.03% to 0.12% on the CNKI dataset).

A single decryption private key is subordinate to multiple users in traditional ciphertext attribute-based encryption schemes, which makes it possible for malicious users or semi-trusted attribute authorities to reveal the decryption private key to third unauthorized parties in order to gain benefits. Moreover, the decryption stage requires numerous pairing calculations, resulting to a huge burden to the end users with limited computing power. To address these issues, this paper proposes an accountable and verifiable outsourced decryption for ciphertext-policy attribute-based encryption. By deploying verifiable outsourcing decryption technology, the majority of the encryption overhead is shifted to the decryption agents , such that the computational load on end users can be reduced. By embedding user identity information and secret information that remains invisible to attribute authority into the user's private key, public accountability of both users and attribute authority is achieved. Security analysis demonstrates that our proposed scheme provides selective security, accountability, and verifiability of outsourcing decryption under the standard model. Performance analysis also indicates that the decryption cost of this scheme mainly lies on the decryption agent side, making it applicable for mobile device users with limited resources.