A reinforcement learning method and apparatus includes storing video frames in a video memory, performing a first preprocessing step of retrieving a sequence of n image frames of the stored video frames, and merging the n image frames in a fading-in fashion by incrementally increasing the intensity of each frame up to the most recent frame having full intensity to obtain a merged frame; and performing a training step of inputting the merged frame to the DQN and training the DQN to learn Q-values for all possible actions from a state represented by the merged frame with only a single forward pass through the network. The learning method and apparatus includes a second preprocessing step of removing the background from the merged frame. The method can be applied to any DQN learning method that uses a convolution neural network as its core value function approximator.

Systems, processes, and computer program products for creating visual designs and arrangements that originate from an image or images are provided. In particular, the present subject matter relates to systems, processes, and computer program products for taking captured images of an intended operating environment and creating visual designs that create visual confusion that can be utilized to disguise a recognizable form of a person or an object by breaking up its outline using portions, magnifications and distortions of a single captured image, portions, magnifications and distortions of multiple captured images, and/or disruptive patterns that can projected on an image screen or can be printed on a material.

A system for performing adaptive focusing of a microscopy device comprises a microscopy device configured to acquire microscopy images depicting cells and one or more processors executing instructions for performing a method that includes extracting pixels from the microscopy images. Each set of pixels corresponds to an independent cell. The method further includes using a trained classifier to assign one of a plurality of image quality labels to each set of pixels indicating the degree to which the independent cell is in focus. If the image quality labels corresponding to the sets of pixels indicate that the cells are out of focus, a focal length adjustment for adjusting focus of the microscopy device is determined using a trained machine learning model. Then, executable instructions are sent to the microscopy device to perform the focal length adjustment.

An image processing apparatus includes a determination unit configured to determine a region of the image on which to perform character recognition processing, a decision unit configured to decide, based on a number of black pixels in contact with the region determined by the determination unit, whether to perform the character recognition processing on an expanded region obtained by expanding the region determined by the determination unit rather than on the region determined by the determination unit, and a character recognition unit configured to perform the character recognition processing on that region of the image decided by the decision unit.

Described herein is a pooling method, device, and system, computer-readable storage medium. The pooling method, comprising: acquiring pixel data of each row where a pooling window is located row by row, each time after the pooling window is moved vertically, wherein the size of the pooling window is NN, N is a positive integer; writing the acquired pixel data of the first N1 rows or a pre-pooling result thereof into a cache; and performing a pooling operation on the pixel data of last row where the pooling window being located and pixel data in the cache, when the pixel data of last row being acquired; outputting a pooling result of the pooling operation. The technical solution of the invention may improve the pooling efficiency and the system performance.

Provided are a method, an apparatus, and a device for identifying human body, including: acquiring a first original picture captured; adjusting a resolution according to the acquired picture to obtain a target picture; processing the target picture based on a preset model for human body feature point detection to determine whether the target picture includes human body information; if the target picture includes the human body information, determining human body area information in the original picture according to the human body information and inputting the human body area information into a filter, enabling that the filter determines target human body area information according to the human body area information; acquiring a next original picture captured; and determining a possible human body area in the next original picture according to the target human body area information, and performing the step of adjusting the resolution according to the possible human body area.

Systems and methods for training image processing neural networks by synthetic photorealistic indicia-bearing images. An example method comprises: generating an initial set of images, wherein each image of the initial set of images comprises a rendering of a text string; producing an augmented set of images by processing the initial set of images to introduce, into each image of the initial set of image, at least one simulated image defect; generating a training dataset comprising a plurality of pairs of images, wherein each pair of images comprises a first image selected from the initial set of images and a second image selected from the augmented set of images; and training, using the training dataset, a convolutional neural network for image processing.

A method and system for determining parameters of an image processing pipeline of a digital camera is disclosed. The image processing pipeline transforms captured image data on a scene into rendered image data. Rendered image data produced by the image processing pipeline of the camera is obtained from the captured image data on the scene. At least a subset of the captured image data on the scene is determined and a ranking order for pixels of the rendered image data is obtained. A set of constraints from the captured image data and the ranked rendered image data is determined, each constraint of the set being determined in dependence on selected pair combinations of pixel values when taken in said ranking order of the rendered image data and corresponding pair combinations of the captured image data. Parameters of the image processing pipeline are determined that satisfy the sets of constraints.

Metadata concerning a photograph is saved onto the photograph in a visible and innocuous way soon after the photograph is captured. The metadata may be generated locally on the imaging device or from a server in communication with the imaging device after analyzing the photograph.

A processor-implemented method implementing a convolution neural network includes: determining a plurality of differential groups by grouping a plurality of raw windows of an input feature map into the plurality of differential groups; determining differential windows by performing, for each respective differential group of the differential groups, a differential operation between the raw windows of the respective differential group; determining a reference element of an output feature map corresponding to a reference raw window among the raw windows by performing a convolution operation between a kernel and the reference raw window; and determining remaining elements of the output feature map by performing a reference element summation operation based on the reference element and each of a plurality of convolution operation results determined by performing respective convolution operations between the kernel and each of the differential windows.