4 results about "Darknet" patented technology

An adaptive, AI-based system for threat analysis of the darknet, consisting of: a) an adaptive network access module configured to establish secure and undetectable connections to Darknet sources using dynamically optimized Tor connections; b) a natural language processing engine specifically developed for cybersecurity, trained on domain-specific datasets to analyze and classify multilingual threat communications; c) a multi-database fusion architecture that integrates document, graph, search and time-series databases through a unified query translation interface; d) a behavioral mimicry system configured to emulate human browsing behavior and randomize browser fingerprints to evade detection by anti-bots; e) a predictive threat intelligence engine that uses machine learning algorithms to predict emerging cyber threats and generate early warnings; f) a distributed crawling infrastructure consisting of containerized crawler instances deployed on geographically distributed, load-balanced servers; and g) an integration and API gateway module configured for secure real-time data exchange with external cybersecurity systems, with all the aforementioned modules communicatively coupled via a central orchestration layer, enabling continuous real-time collection, analysis, correlation and prediction of dark web threats.

The invention relates to an animal attitude estimation method based on key point perception enhancement, and belongs to the field of computer vision. The method comprises the following steps: constructing and training an animal posture estimation network RCBPose; wherein the RCBPose comprises an improved Darknet-53 backbone network, a CSConv module, a Neck network and a Head part, and the improved Darknet-53 backbone network, the CSConv module, the Neck network and the Head part are connected in sequence; performing feature extraction on an image to be subjected to animal attitude estimation through the improved Darknet-53 backbone network to obtain a high-dimensional feature map; performing convolution processing on the high-dimensional feature map through a CSConv module, and outputting a multi-scale feature map; performing feature fusion on the multi-scale feature map through a Neck network to obtain fusion features including animal key point positions and global position information; and the fusion features are identified and predicted through the Head part to obtain spatial position information between animal key points, and animal attitude estimation is completed. The objective of the invention is to solve the technical problem of inaccurate animal posture and key point information caused by shielding and background interference in a complex farm environment in the prior art.

This invention relates to the fields of computer vision and autonomous driving perception technology, and discloses a method for target detection in adverse weather conditions. Addressing the shortcomings of existing technologies such as insufficient real-time performance, weak noise suppression, and poor multi-scale perception capabilities, an AWR-YOLO end-to-end framework is constructed. This framework uses an enhanced CSP-Darknet as its backbone, embeds MS-Block to expand the effective receptive field, and deploys LAS strategies (including GSConv and GDAM) to correct geometric distortion and suppress noise. A selective replacement strategy is employed to optimize deployment, achieving a balance between accuracy and efficiency. Experiments show that the model achieves an mAP50 of 68.89%, an mAP50-95 of 55.69%, and an inference speed of 118.5 FPS, demonstrating excellent robustness and suitability for autonomous driving edge devices.

The application relates to an animal posture estimation method based on key point perception enhancement and belongs to the field of computer vision. The method comprises the following steps: constructing and training an animal posture estimation network RCBPose; wherein the RCBPose comprises an improved Darknet-53 backbone network, a CSConv module, a Neck network and a Head part; the improved Darknet-53 backbone network is used for performing feature extraction on an image to be subjected to animal posture estimation, so as to obtain a high-dimensional feature map; the CSConv module is used for performing convolution processing on the high-dimensional feature map, so as to output a multi-scale feature map; the Neck network is used for performing feature fusion on the multi-scale feature map, so as to obtain fused features containing animal key point positions and global position information; and the Head part is used for identifying and predicting the fused features, so as to obtain spatial position information between animal key points and complete animal posture estimation. The application aims to solve the technical problem that, in the prior art, animal posture and key point information are inaccurate due to occlusion and background interference in a complex breeding farm environment.