Recently, the application of deep reinforcement learning into field session-based interactive recommendation has attracted great attention from researchers. However, despite that some models based on have been proposed, they still suffer to following limitations: (1) these works ignore skip behaviors sequential patterns in users' clicking behavior; (2) fail incorporate positive feedback and negative proposed recommender system when is sparse. Therefore, solve problems mentioned above, a...

With the development of autonomous perception and obstacle avoidance technologies for power lines inspection using unmanned aerial vehicles(UAV), methods extraction distance measurement from binocular images become spotlights gradually. In this paper, Ratio detection operator fast Hough transform are adopted to extract pixels detect straight respectively, a pair selected by demand-based strategy further. Then, aim at lines, method constructing homonymy points based on linear hypothesis line...

Many prediction tasks of real-world applications need to model multi-order feature interactions in user's event sequence for better detection performance. However, existing popular solutions usually suffer two key issues: 1) only focusing on and failing capture the influence; 2) information, but ignoring internal relations each event, thus extract a representation. In this paper, we consider two-level structure capturing hierarchical information over sequence: learning effective based...

Nowadays, unmanned aerial vehicles (UAVs) have achieved massive improvement, which brings great convenience and advantage. Meanwhile, threats posed by them may damage public security personal safety. This article proposes an architecture of intelligent anti-UAVs low-altitude defense system. To address the key problem discovering UAVs, research based on multisensor information fusion is carried out. Firstly, to solve probing suspicious targets, a method designed, combines radar photoelectric...

In the research of interactive music generation, we propose a generation method, that computer generates music, under recognition human conductor's gestures. this research, generated is tuned by recognized gestures for parameters network chaotic elements in realtime. To make outcomes more closer to different method proposed melody specifiable. Music theories are embedded algorithm, as result, will be richer. Furthermore, reconstructed system and performed experiment generating composed human.

In order to enhance safety and improve the multimedia wireless sensor information's confidentiality information processing speed, research of data encryption chip core is proposed. Under action key, can encrypt plaintext decrypt cipher text from input port properly. The key must be randomly distributed. First, this paper analyzes encrypted architecture. Second, it introduces design principle chip, randomness. Finally, work function designed nuclear verified by experiments. correctly....

New fast algorithm has be brought forward for the linear congruence generator such as y(n+1)=(16807 × y(n))mod(2^31-1) and realized. The calls one time 32-bit multiplier; two-times adders, a few bit shifters highest segregator, with excluding subtraction division. experiment shows that is 40 percent more than other algorithm. realized in hard describing language three registers.

With its miniaturization, intelligence, popularization and diversification, drones have become one of the most rapidly developing emerging industries in recent years. However, illegal use leads to more threats key areas, there are problems such as difficult detection lack effective means interception. In view targets, methods defense systematically sorted out. Based on advantages disadvantages each equipment, this paper puts forward an important anti-drone architecture for prevention...

With the energy consumption, new route may be reconstructed with sufficient nodes. In order to evaluate of nodes, an consumption model was raised calculate node in wireless sensor networks. The nodes setup data structure, rule and radio networks radio. simulation algorithm designed realized. result shows that can simulate for which is helpful algorithm.

Despite demonstrating superior rate-distortion (RD) performance, learning-based image compression (LIC) algorithms have been found to be vulnerable malicious perturbations in recent studies. Adversarial samples these studies are designed attack only one dimension of either bitrate or distortion, targeting a submodel with specific ratio. However, adversaries real-world scenarios neither confined singular dimensional attacks nor always control over ratios. This variability highlights the...

The research on intent-enhanced sequential recommendation algorithms focuses how to better mine dynamic user intent based behavior data for tasks. Various augmentation methods are widely applied in current algorithms, effectively enhancing the ability capture intent. However, these used often rely a large amount of random sampling, which can introduce excessive noise into training data, blur intent, and thus negatively affect performance. Additionally, have limited approaches utilizing...

Video traffic surveillance is of high interest in the field intelligent transportation systems and moving vehicle tracking an essential technique. Particle filter approximate optimal Bayesian solution for as a nonlinear or non-Gaussian system. In this paper method based on PF presented, which combines gray contour feature particles using fusion algorithm to balance weights according present scene. It adaptable scene because it utilizes advantage proper The experiments demonstrate that...

With the rapid development of medical imaging technologies, high-resolution CT image data is great value for research as well a clinical diagnosis. The paper takes lung an example. Retrieving images similar to input one helpful assist physicians in Compared with traditional content-based retrieval, similarity retrieval requires higher accuracy, requirements external shape internal vascular and lesion location similarity. In state-of-the-art supervised deep learning networks, network relies...