Abstract: A data processing method performed by a first device includes: generating, in response to first service data satisfying a data uploading condition, a first bit array corresponding to the first service data; encrypting the first bit array through a data key to obtain a ciphertext bit array, the data key being generated by a second device in a data intersection application run in a trusted execution environment of the second device; and transmitting the ciphertext bit array to a blockchain node for forwarding to a second device, for the second device to decrypt, in the data intersection application through the data key, the ciphertext bit array to obtain the first bit array.

Abstract: A vehicle hydraulic control system comprises a high-low pressure decoupling valve, a high-pressure control oil circuit including a main pressure regulating pilot solenoid valve connected to a transmission controller and used for regulating the main oil pressure according to the instructions of the transmission controller and a main pressure regulating mechanical valve, a low-pressure cooling and lubricating oil circuit, a mechanical pump communication with the high-pressure control oil circuit, and an electronic pump is in communication with the high-pressure control oil circuit or the low-pressure cooling and lubricating oil circuit by means of the high-low pressure decoupling valve. The main pressure regulating mechanical valve is used for guiding the flow output by the electronic pump and/or the mechanical pump in communication with the high-pressure control oil circuit to the low-pressure cooling and lubricating oil circuit according to the control of the main pressure regulating pilot solenoid valve.

Abstract: An acceleration and compression method for a deep convolutional neural network based on quantization of a parameter provided by the present application comprises: quantizing the parameter of the deep convolutional neural network to obtain a plurality of subcode books and respective corresponding index values of the plurality of subcode books; acquiring an output feature map of the deep convolutional neural network according to the plurality of subcode books and respective corresponding index values of the plurality of subcode books. The present application may implement the acceleration and compression for a deep convolutional neural network.

Abstract: An acceleration and compression method for a deep convolutional neural network based on quantization of a parameter provided by the present application comprises: quantizing the parameter of the deep convolutional neural network to obtain a plurality of subcode books and respective corresponding index values of the plurality of subcode books; acquiring an output feature map of the deep convolutional neural network according to the plurality of subcode books and respective corresponding index values of the plurality of subcode books. The present application may implement the acceleration and compression for a deep convolutional neural network.

Abstract: An image matching method based on cascaded binary encoding includes using a hashing look-up with multiple hashing tables to coarsely filter candidate key-points in an image to produce a candidate subset of key-points, projecting the candidate subset into a high-dimensional Hamming space, and building a “Hamming distance-memory address” hashing table. An optimal matching key-point is discovered by querying the hashing table. The image matching method has high processing speed and matching quality, which can be used for efficient and accurate image matching.

Abstract: This invention involves an image matching method based on the cascaded binary encoding. The stated method includes: Procedure S1, using the hashing look-up with multiple hashing tables to coarsely filter candidate key-points in the image to produce a candidate subset of key-points; Procedure S2, projecting the candidate subset into a high-dimensional Hamming space; Procedure S3, a “Hamming distance-memory address” hashing table is built, and the optimal matching key-point is discovered by querying this hashing table. The image matching method proposed in this invention has high processing speed and matching quality, which can be used for efficient and accurate image matching.

Abstract: A method and system for invoking an operation of a communication terminal in response to registering and interpreting a predetermined motion or pattern of an object. An input is received, the image data of the object is captured and the object in the image data is identified.

Abstract: A method and system for invoking an operation of a communication terminal in response to registering and interpreting a predetermined motion or pattern of an object. An input is received, the image data of the object is captured and the object in the image data is identified.