Search

The traditional rearview mirror method cannot fully guarantee safety when driving trucks. RGB and infrared images collected by cameras are used for registration and recognition, so as to achieve the perception of surroundings and ensure safe driving. The traditional scale-invariant feature transform (SIFT) algorithm has a mismatching rate, and the YOLO algorithm has an optimization space in feature extraction. To address these issues, this paper proposes a truck surround sensing technique based on multi-features and an improved YOLOv5 algorithm. Firstly, the edge corner points and infrared features of the preset target region are extracted, and then a feature point set containing the improved SIFT algorithm is generated for registration. Finally, the YOLOv5 algorithm is improved by fusing infrared features and introducing a composite prediction mechanism at the prediction end. The simulation results show that, on average, the image stitching accuracy is improved by 17%, the time is reduced by 89%, and the target recognition accuracy is improved by 2.86%. The experimental results show that this method can effectively perceive the surroundings of trucks, accurately identify targets, and reduce the missed alarm rate and false alarm rate. [ABSTRACT FROM AUTHOR]

Private set union (PSU) allows several parties to obtain the union of their private sets without disclosing each party's private information. Existing PSU protocols often have polynomial complexity for the complete set size or complicated process. In this paper, a quantum multi-party PSU protocol based on least common multiple (LCM) and Shor's algorithm is proposed, which enables the union of multiple sets to be computed all at once. In order to increase the one-time success probability of the protocol, we first improved Shor's period-finding algorithm, which is used in LCM computation and integer factoring. Each party's private set is encoded into an integer obtained by multiplying several prime numbers, thus the PSU problem is transformed into an LCM problem. The LCM of these integers is computed by using the improved Shor's period-finding algorithm, and then factored to derived the union set. We prove the correctness of the proposed protocol, and its unconditional security against semi-honest attacks. Complexity analysis shows that our protocol has logarithmic complexity for the complete set size. [ABSTRACT FROM AUTHOR]

Classical forgery attacks against Offset Two-round (OTR) structures require some harsh conditions, such as some plaintext and ciphertext pairs need to be known, and the success probability is not too high. To solve these problems, a quantum forgery attack on OTR structure using Simon's algorithm is proposed. The attacker intercepts the ciphertext-tag pair (C , T) between the sender and receiver, while Simon's algorithm is used to find the period of the tag generation function in OTR, then we can successfully forge new ciphertext C ′ ( C ′ ≠ C) for intercepted tag T. For a variant of OTR structure (Prøst-OTR-Even-Mansour structure), a universal forgery attack, in which it is easy to generate the correct tag of any given message if the attacker is allowed to change a single block in it, is proposed. It first obtains the secret parameter L using Simon's algorithm, then the secret parameter L is used to find the keys k 1 and k 2 , so that an attacker can forge the changed messages. It only needs several plaintext blocks to help obtain the keys to forge any messages. Performance analysis shows that the query complexity of our attack is O (n) , and its success probability is very close to 1. [ABSTRACT FROM AUTHOR]

Exploiting quantum mechanisms, quantum attacks have the potential ability to break the cipher structure. Recently, Ito et al. proposed a quantum attack on Feistel-2* structure (Ito et al.'s attack) based on the Q2 model. However, it is not realistic since the quantum oracle needs to be accessed by the adversary, and the data complexity is high. To solve this problem, a quantum all-subkeys-recovery (ASR) attack based on multi-equations quantum claw-finding is proposed, which takes a more realistic model, the Q1 model, as the scenario, and only requires 3 plain-ciphertext pairs to quickly crack the 6-round Feistel-2* structure. First, we proposed a multi-equations quantum claw-finding algorithm to solve the claw problem of finding multiple equations. In addition, Grover's algorithm is used to speedup the rest subkeys recovery. Compared with Ito et al.'s attack, the data complexity of our attack is reduced from O (2 n) to O(1), while the time complexity and memory complexity are also significantly reduced. [ABSTRACT FROM AUTHOR]