6results about How to "Avoid attack" patented technology

The application discloses a method for establishing end-to-end secure communication of unmanned aerial vehicle under cloud network architecture, and takes cloud server as a security service center to provide full-process authentication service for unmanned aerial vehicle end system; the cloud server only issues public authentication materials to unmanned aerial vehicles, and session key negotiation and establishment of end-to-end secure communication are realized between unmanned aerial vehicles; the unmanned aerial vehicle end system does not need to store authentication materials of other unmanned aerial vehicles in advance, the cloud server is driven by specific tasks, and specific authentication materials required for key negotiation are issued to unmanned aerial vehicles which need to be securely cooperated according to task requirements. The application only requires that the cloud server and any one of the unmanned aerial vehicles for key negotiation establish a communication connection, and the authentication materials issued by the cloud server can make the unmanned aerial vehicles realize key negotiation and establish an end-to-end secure channel in a specific range; different temporary public keys are used for each request message, which can effectively prevent the harm caused by long-term and short-term secret leakage.

This application provides a face recognition method, electronic device, storage medium, and program product. The method includes: acquiring a first image in response to a face recognition operation; processing the first image to generate a first number of region images, the first number of region images having the same size but different dimensional information, the dimensional information representing the detection scale; inputting the first number of region images into a first model to generate multi-scale feature data, the multi-scale feature data representing the global features of the first image at different detection scales; processing the multi-scale feature data through a second model to generate deep feature data, the deep feature data representing the deep features of the first image at different detection scales; and obtaining a face recognition result based on the deep feature data. This method aims to improve the accuracy of real / fake face discrimination using a liveness detection model and prevent various liveness attacks.

The application discloses a kind of based on unified port's SIP and RTP tunnel encryption forwarding method and system.The method is cooperated by tunnel client deployed in client side and tunnel server side of server side, including: based on unified port, establishing encrypted communication tunnel between the two;Tunnel client receives SIP signaling, modifies media connection information in it, and then sends to server side through tunnel encryption, and is forwarded to SIP server by server side;Tunnel client receives RTP media stream, and then sends to server side through tunnel encryption after additional header containing forwarding destination information is encapsulated, and is forwarded to SIP server according to header information by server side.The application tunnels and encrypts transmission of SIP signaling and RTP media stream through single port, effectively avoids network protocol ban, greatly simplifies firewall configuration, significantly improves NAT penetration success rate and communication security.

This application discloses an image recognition method, storage medium, and device, which can be applied to the fields of artificial intelligence, intelligent transportation, and assisted driving. The method includes: acquiring an image to be recognized, including a target object; extracting planar features from the image to be recognized to obtain planar feature information of the target object; converting the image to be recognized into a depth image; extracting depth features from the depth image to obtain depth feature information of the target object; fusing the planar feature information and the depth feature information to obtain fused feature information of the target object; and identifying the attributes of the target object based on the fused feature information to obtain attribute information of the target object. This application can, to a certain extent, prevent attacks on image feature information and improve the reliability and accuracy of image recognition.

This application discloses an image recognition method, storage medium, and device, which can be applied to the fields of artificial intelligence, intelligent transportation, and assisted driving. The method includes: acquiring an image to be recognized, including a target object; extracting planar features from the image to be recognized to obtain planar feature information of the target object; converting the image to be recognized into a depth image; extracting depth features from the depth image to obtain depth feature information of the target object; fusing the planar feature information and the depth feature information to obtain fused feature information of the target object; and identifying the attributes of the target object based on the fused feature information to obtain attribute information of the target object. This application can, to a certain extent, prevent attacks on image feature information and improve the reliability and accuracy of image recognition.