A learning data generation device that can generate learning data suitable for learning of an identification model. The learning data generation device has a function of cutting out part of first image data as second image data, a function of generating a two-dimensional graphic corresponding to the area of the second image data and representing a pseudo defect, a function of generating third image data by combining the second image data and the two-dimensional graphic, and a function of assigning a label corresponding to the two-dimensional graphic to the third image data. By using the third image data for learning of the identification model, a highly accurate identification model can be generated.

A method includes: computing noise data by subtracting, by a processing circuit, a noisy image from a corresponding ground truth image; clustering, by the processing circuit, a plurality of noise values of the noise data based on intensity values of the corresponding ground truth image; permuting, by the processing circuit, a plurality of locations of the noise values of the noise data within each cluster; generating, by the processing circuit, a synthetic noise image based on the permuted locations of the noise values; adding, by the processing circuit, the synthetic noise image to the corresponding ground truth image to generate a synthetic noisy image; and augmenting an image dataset for training a neural network to perform image denoising with the synthetic noisy image.

Object detection is performed for an image acquired by imaging using an array sensor in which a plurality of imaging elements are arranged one-dimensionally or two-dimensionally, some of the imaging elements are configured as color-filter-disposed pixels in which a color filter is disposed in an incident optical path, and color information acquisition points are formed by the color-filter-disposed pixels. Then, coloring processing in a pixel range of a detected object is performed by referring to color information acquired at the color information acquisition points corresponding to the inside of the pixel range of the detected object.

A tree crown extraction method based on UAV multi-source remote sensing includes: obtaining a visible light image and LIDAR point clouds, taking a digital orthophoto map (DOM) and the LIDAR point clouds as data sources, using a method of watershed segmentation and object-oriented multi-scale segmentation to extract single tree crown information under different canopy densities. The object-oriented multi-scale segmentation method is used to extract crown and non-crown areas, and a tree crown distribution range is extracted with the crown area as a mask; a preliminary segmentation result of single tree crown is obtained by the watershed segmentation method based on a canopy height model; a brightness value of DOM is taken as a feature, the crown area of the DOM is performed secondary segmentation based on a crown boundary to obtain an optimized single tree crown boundary information, which greatly increases the accuracy of remote sensing tree crown extraction.

A mapping curve parameter obtaining method and apparatus are described. The method includes obtaining a first mapping curve parameter set and first maximum target system display luminance, and obtaining a display luminance parameter set, where the display luminance parameter set includes maximum display luminance and/or minimum display luminance of a display device. The method also includes obtaining an adjustment coefficient set, where the adjustment coefficient set includes one or more adjustment coefficients, and the one or more adjustment coefficients correspond to one or more parameters in the first mapping curve parameter set. Furthermore, the method includes adjusting the one or more parameters in the first mapping curve parameter set based on the display luminance parameter set, the first maximum target system display luminance, and the adjustment coefficient set to obtain a second mapping curve parameter set, where the second mapping curve parameter set includes one or more adjusted parameters.

A handwash monitoring system includes: an imaging device; and a processor. The processor detects a first candidate abnormality existing in a hand of a user from a first image captured by the imaging device before handwashing, and detects a second candidate abnormality existing in the hand of the user from a second image captured by the imaging device after the handwashing. The processor determines a type of an abnormality on the hand of the user based on a difference between a shape of the first candidate abnormality and a shape of the second candidate abnormality wherein the first candidate abnormality and the second candidate abnormality are detected from an identical region.

Disclosed are an image processing method and apparatus, a computer device and a storage medium, which relate to the field of artificial Intelligence. The method includes receiving an original image; performing image photographing defect detection and color deviation detection on the original image, the image photographing defect detection determining whether there exists a photographing defect that is irreparable through image processing, the color deviation detection determining whether there exists color cast in the original image; performing color correction on the original image if the original image passes the image photographing defect detection and does not pass the color deviation detection; and generating a target image based on the color-corrected original image.

A method is specified for determining a size of a defect occurring in a surface region of a component while a surface modification process is performed on the surface region. The method includes identifying an occurrence of a defect occurring in a surface region of a component on a basis of a set of images and determining a size of the defect in a separate method step from the occurrence of the defect identified. In addition, an apparatus and a computer program are specified for determining a size of a defect occurring in a surface region of a component while a surface modification process is performed on the surface region.

Systems and methods for image-based perception. The methods comprise: capturing images by a plurality of cameras with overlapping fields of view; generating, by a computing device, spatial feature maps indicating locations of features in the images; identifying, by the computing device, overlapping portions of the spatial feature maps; generating, by the computing device, at least one combined spatial feature map by combining the overlapping portions of the spatial feature maps together; and/or using, by the computing device, the at least one combined spatial feature map to define a predicted cuboid for at least one object in the images.

Provided are an apparatus for acquiring a depth image, a method for fusing depth images, and a terminal device. The apparatus for acquiring a depth image includes an emitting module, a receiving module, and a processing unit. The emitting module is configured to emit a speckle array to an object, where the speckle array includes p mutually spaced apart speckles. The receiving module includes an image sensor. The processing unit is configured to receive the pixel signal and generate a sparse depth image based on the pixel signal, align an RGB image at a resolution of a*b with the sparse depth image, and fuse the aligned sparse depth image with the RGB image using a pre-trained image fusion model to obtain a dense depth image at a resolution of a*b.