Abstract: Systems and methods for performing optical distance measurement are provided. In one aspect, a system for measuring a distance to an object comprises a light emitter configured to emit an outbound light pulse, and a light sensor configured to receive a returning light pulse reflected from the object and output an analog pulse signal representing the returning light pulse. The system also comprises a field-programmable gate array (FPGA) coupled to the light sensor. The FPGA is configured to convert the analog pulse signal to a plurality of digital signal values, and generate a plurality of time measurements corresponding to the plurality of digital signal values. The system also comprises a controller configured to calculate the distance to the object based on the plurality of digital signal values and the plurality of time measurements.

Abstract: A method for controlling an aerial vehicle includes

Abstract: A method for processing flight data includes acquiring, by detecting devices carried by one or more first aerial vehicles, environmental data associated to a position region; and sending, by the one or more first aerial vehicles, the environmental data to an external device. The environmental data is configured to generate a flight control instruction for controlling an operation of a second aerial vehicle while the second aerial vehicle is within the position region.

Abstract: A time measurement correction method includes a field programmable gate array (FPGA) determining to enter a self-correction mode of time measurement, and in the self-correction mode, the FPGA controlling to generate a standard signal and the FPGA controlling to obtain the standard signal and collecting measurement data of at least one TDC channel included in the FPGA based on the standard signal. The standard signal is used to correct the at least one TDC channel of the FPGA.

Abstract: Described herein are systems, methods, storage media, and computer programs for tracking a plurality of feature points in a first image and a second image. In one embodiment, a first layer of a first image pyramid for the first image and a first layer of a second image pyramid for the second image are obtained. Based on expected pixel locations of the plurality of feature points in the first layer of the second image pyramid, a feature point is selected from the plurality of feature points. For the feature point, a first pixel in the first layer of the first image pyramid is determined, and a second pixel in the first layer of the second image pyramid is determined, where the first pixel substantially matches the second pixel.

Abstract: A method for processing flight data includes acquiring, by detecting devices carried by one or more first aerial vehicles, environmental data associated to a position region; and sending, by the one or more first aerial vehicles, the environmental data to an external device. The environmental data is configured to generate a flight control instruction for controlling an operation of a second aerial vehicle while the second aerial vehicle is within the position region.

Abstract: Convolutional neural network-based image processing method and device are provided. The device includes a first on-chip memory and an arithmetic circuit configured to read a 3D feature map from a first on-chip memory by blocks the 3D feature map being divided into L blocks, perform processing of the current layer of the convolutional neural network on the 3D feature map by blocks; and store an output result of the current layer to the first on-chip memory. The first on-chip memory includes: S first storage spaces, each being used to store one of the L blocks included in the 3D feature map as input data of the current layer; and R second storage spaces, each being used to store output data of the current layer of one of the L blocks. L, S and R are integers greater than 1, and S and R are less than L.

Abstract: The present disclosure provides an optical flow tracking device. The device includes an image pyramid establishing circuit configured to establish a first image pyramid of a first image and a second image pyramid of a second image. The first image pyramid includes one or more image layers, and the second image pyramid includes one or more image layers. The device further includes an optical flow tracking circuit configured to perform an optical flow tracking processing on a kth image layer in the first image pyramid and a kth image layer in the second image pyramid to acquire an optical flow estimator of the second image in the kth image layer. K is an integer. The optical flow tracking circuit is configured to calculate an optical flow of the second image by using the optical flow estimator of the second image in the kth image layer.

Abstract: A data processing method includes reading, from a first storage unit into a second storage unit, one or more entire layers of data at a top of one or more image pyramids having a plurality of layers; computing, according to to-be-processed data at each layer stored in the second storage unit, a processing result at each layer stored in the second storage unit; and reading and processing, in sequence, data at one or more remaining layers of the one or more image pyramids from the first storage unit. The one or more image pyramids include a first original pyramid of a first image and a second original pyramid of a second image.

Abstract: One or more block configurations are selected, based on object information associated with an object, from a plurality of block configurations having at least one of different sizes or different shapes. A first point in a first image is matched with a corresponding second point in a second image using the one or more block configurations.

Abstract: A pooling device includes one or more first processing circuits and one or more second processing circuits. The one or more first processing circuits are configured to compute temporary pooling results of an input image along a row direction or a column direction. The one or more second processing circuits are configured to generate an output image according to the temporary pooling results of the input image along the row direction or the column direction.

Abstract: A method for controlling an aerial vehicle includes obtaining flight indication data containing a position region associated with the flight indication data, determining whether a current position of the aerial vehicle is within the position region associated with the flight indication data, and, in response to the current position being within the position region, generating a flight control instruction according to the flight indication data, and controlling a flight of the aerial vehicle according to the flight control instruction.

Abstract: A data processing method includes reading first to-be-processed data from a first storage unit into a second storage unit, computing a processing result according to the first to-be-processed data stored in the second storage unit, and, while computing the processing result, reading second to-be-processed data from the first storage unit. The first to-be-processed data is at a first layer of one or more image pyramids having a plurality of layers, and the second to-be-processed data is at a second layer of the one or more image pyramids.

Abstract: A method of block matching includes selecting a block configuration from a plurality of block configurations and matching a first point in a first image with a corresponding second point in a second image using the selected block configuration.

Abstract: A connected component detection method includes reading an image data block from an external address block of an external storage medium, storing the read image data block in an internal address block of an internal storage medium, and performing a connected component detection using the image data block stored in the internal address block. The image data block is one of a plurality of image data blocks obtained by dividing original image data.

Abstract: A method for processing across neural networks includes: when processing a last block of a plurality of blocks of an i-th layer of a first neural network, reading data of a first block of a plurality of blocks of a k-th layer of a second neural network from a memory; and processing the first block of the plurality of blocks of the k-th layer of the second neural network according to the data of the first block of the plurality of blocks of the k-th layer of the second neural network after processing the last block of the plurality of blocks of the i-th layer of the first neural network. 1?i?N, N is a number of layers of the first neural network; and 1?k?M, M is a number of layers of the second neural network.

Abstract: Representative embodiments of the present technology include a device for measuring distance to an object. The device comprises a light emitter configured to emit an outbound light pulse and a light sensor configured to receive a returning light pulse and output a pulse signal representing the returning light pulse. The device further comprises a field-programmable gate array (FPGA) coupled to the light sensor and including a time-to-digital converter (TDC) having a series of sequentially coupled delay units. Individual sequentially coupled delay units are associated with corresponding individual delay times. At least some of the sequentially coupled delay units have different individual delay times. The TDC is configured to measure timing information of the pulse signal based at least in part on the individual delay times of the sequentially coupled delay units. The device further includes a controller configured to calculate the distance to the object based on the timing information.

Abstract: An image processing method comprising acquiring a plurality of pixel rows of an image according to a rectangular pixel region, and storing the plurality of pixel rows as blocks in a plurality of line buffers. A pixel row is stored as a block in a storage space of the plurality of line buffers, and a number of bits of the block in a line buffer is smaller than a number of pixel columns of the image.

Abstract: The present disclosure relates to a computation apparatus for a neural network. The computation apparatus includes a first processing unit and a second processing unit. The first processing unit is configured to perform a first computation on k1 number of input feature data according to a size of a computation window to obtain an intermediate result, where a size of the computation window is k1×k2, and k1 and k2 are positive integers. The second processing unit is configured to perform a second computation on k2 number of intermediate results output by the first processing unit according to the size of the computation window to obtain a computation result.

Abstract: A time measurement circuit includes a signal input configured to receive a to-be-tested signal, a delay line including n delay units that are sequentially connected and include a first delay unit connected to the signal input to receive the to-be-tested signal, a logic controller including an input connected to the signal input to receive the to-be-tested signal and an output connected to a k-th delay unit of the n delay units, and a latch connected to the n delay units and configured to latch output signals of the n delay units. n is a positive integer larger than two, and k is a positive integer larger than one and smaller than n.