Sony Corporation

Chinese Academy of Sciences

Xiong Yin

Liang Li

Kun Liu

Zhongxing Wang

Lili Kang

Deng Zhang

Zhiyao Liu

Ruo Wang

Zhiguo An

A semi-airborne electromagnetic survey device and a semi-airborne electromagnetic survey method based on coaxial coplanar mutual reference coil assembly are provided. The device includes a measuring coil and a reference coil with linear correlation of additive motion noise; the reference coil and the measuring coil have a same bandwidth, are coaxial and coplanar and are in hard connection or soft connection; detection resolution of the reference coil only distinguishes the additive motion noise; detection resolution of the measuring coil distinguishes real vertical magnetic field signals and motion noise simultaneously; the reference coil has a much smaller outer diameter than the measuring coil. In the method, one coil only receives the additive motion noise, and an other coil simultaneously receives the additive motion noise and the real vertical magnetic field signals; and the additive motion noise received by the two coils is cancelled to obtain the real vertical magnetic field signals.

Semi-airborne electromagnetic survey device and method based on coaxial coplanar mutual reference coil assembly

Disclosed is a magnetotelluric inversion method based on a fully convolutional neural network. The magnetotelluric inversion method includes: constructing a multi-dimensional geoelectric model; constructing a fully convolutional neural network structure model to obtain initialized fully convolutional neural network model parameters; training and testing the fully convolutional neural network structure model based on the training sets and the test sets to obtain optimized fully convolutional neural network structure model parameters; determining whether training of the fully convolutional neural network structure model is completed according to fitting error changes corresponding to the training sets and the test sets; and finally, inputting measured apparent resistivity into a trained fully convolutional neural network structure model for inversion, and further optimizing the fully convolutional neural network structure model by analyzing precision of an inversion result until an inversion fitting error satisfies a set error requirement.

Magnetotelluric inversion method based on fully convolutional neural network

The disclosure relates to a monitoring system and a control method for a monitoring system. The monitoring system includes a camera device, an object recognizing device and an arrangement control device. The camera device includes at least one movable camera device. The object recognizing device is configured to recognize a target object based on a primary image captured by the camera device. The arrangement control device is configured to control the spatial arrangement of the movable camera device so that the movable camera device is located at a predetermined position relative to the target object.

Monitoring system and control method thereof

The disclosure relates to an image processing method and image processing device for a monitoring system including multiple camera devices, and a monitoring system. The image processing method includes: extracting a key frame from a video acquired by each of the plurality of camera devices; for each of the extracted key frames, determining a neighbor camera device having monitor area adjacent to the monitor area of the key frame, and acquiring from the neighbor camera device a corresponding video frame which is simultaneous to the key frame; blending the key frame extracted from each of the camera devices and the corresponding video frame to generate a panorama key frame; and generating a sequence formed by panorama key frames for the key frames as a video abstract of the monitoring system.

Growing community of inventors

Beijing, China

Xiong Yin