Abstract: Embodiments of this application disclose a method and apparatus for training an image region segmentation model, and an image region segmentation method and apparatus. The method includes acquiring a sample image set, and each image of the sample image set having first annotation information; generating graph structure data corresponding to a sample image in the sample image set, the graph structure data comprising multiple nodes, and each node comprising at least one pixel in the sample image; determining second annotation information of each node according to the graph structure data and the first annotation information corresponding to the sample image by using a graph convolutional network model, a granularity of the second annotation information being smaller than a granularity of the first annotation information, the graph convolutional network model being a part of an image segmentation model; and training the image segmentation model according to the second annotation information.

Abstract: This application relates to a video compression and video decompression that includes: generating, based on a key frame and a forward search frame of a video, a pixel kernel of each pixel in the key frame; performing smoothness constraint processing on the pixel kernel, to obtain a target pixel kernel; compressing the key frame and the target pixel kernel, to obtain a compressed key frame and a compressed pixel kernel; compressing a residual graph between the forward search frame and a predicted frame, to obtain a compressed graph, the predicted frame being a video frame generated based on the target pixel kernel and the key frame; and obtaining a compressed video packet according to the compressed graph, the compressed key frame, and the compressed pixel kernel.

Abstract: This application discloses a video compression method, a video decoding method, and related apparatuses. The method includes: extracting a key point from a to-be-processed video frame and a previous video frame respectively to obtain first position information and second position information; performing motion estimation based on the first position information and the second position information to obtain motion information; performing image inpainting based on the motion information and the previous video frame to obtain an initial video frame; determining a latent feature based on the to-be-processed video frame and the initial video frame; and performing video compression based on the first position information, the second position information, and the latent feature to obtain a video compressed file.

Abstract: The disclosure provides an image processing method and apparatus. The method may include acquiring a target image having at least two division regions. The division region includes an annotation point, and the annotation point is corresponding to a target segmentation object. The method may further include performing iterative expansion on the annotation points to obtain an iterative expansion region. The method may further include determining a first loss result between the iterative expansion region and the division region and a second loss result between the iterative expansion region and the annotation point. The method may further include, when the first loss result and the second loss result meeting a termination condition, stopping the iterative expansion to obtain a resulting iterative expansion region. The method may further include determining the resulting iterative expansion region as a region where the target segmentation object is located in the target image.

Abstract: The present disclosure relates to a method for training a neural network model performed at an electronic device. The method includes: performing initial training by using a first training sample set to obtain an initial neural network model; performing a prediction on a second training sample set by using the initial neural network model to obtain a prediction result of each of training samples in the second training sample set; determining a plurality of preferred samples from the second training sample set based on the prediction results; adding the plurality of preferred samples that are annotated to the first training sample set to obtain an expanded first training sample set; updating training of the initial neural network model by using the expanded first training sample set to obtain an updated neural network model until a training ending condition is satisfied.

Abstract: A medical image processing method includes: obtaining a biological tissue image including a biological tissue, recognizing, in the biological tissue image, a first region of a lesion object in the biological tissue; recognizing a lesion attribute matching the lesion object; dividing an image region of the biological tissue in the biological tissue image into a plurality of quadrant regions; obtaining target quadrant position information of a quadrant region in which the first region is located; and generating medical service data according to the target quadrant position information and the lesion attribute.

Abstract: This application discloses a physiological image processing method performed by a computer device. The method includes: obtaining a physiological image; determining position information of at least one mutated object in the physiological image based on a physiological image processing model; performing color channel decomposition on the physiological image to obtain staining information corresponding to the physiological image; and making statistics according to the position information and the staining information to obtain a staining counting result of the mutated objects. According to this application, by obtaining the staining counting result of the mutated objects according to the position information of the mutated objects and the staining information, the staining counting result of the mutated objects is quickly obtained.

Abstract: A medical image region screening method and apparatus and a storage medium are provided. The method includes: obtaining a medical image of biological tissue, segmenting tissue regions of a plurality of tissue types from the medical image, selecting, from the tissue regions of the plurality of tissue types based on types of capturing positions of the medical image, a reserved region, obtaining a positional relationship between the reserved region and a predicted lesion region in the medical image; and screening for the predicted lesion region in the medical image based on the positional relationship, to obtain a target lesion region.

Abstract: A medical image processing method includes: obtaining a biological tissue image including a biological tissue, recognizing, in the biological tissue image, a first region of a lesion object in the biological tissue; recognizing a lesion attribute matching the lesion object; dividing an image region of the biological tissue in the biological tissue image into a plurality of quadrant regions; obtaining target quadrant position information of a quadrant region in which the first region is located; and generating medical service data according to the target quadrant position information and the lesion attribute.

Abstract: A computer device, obtains a mammographic image of a unilateral breast. The mammographic image includes a cranial-caudal (CC)-position mammographic image and a mediolateral-oblique (MLO)-position mammographic image. The computer device invokes a breast detection model to perform a prediction of a condition of the unilateral breast according to the CC-position mammographic image and the MLO-position mammographic image. The device obtains a prediction result of the unilateral breast, and generates and outputs a detection report that includes the prediction result.

Abstract: A hydraulic rotary side-swing elevator includes an elevator body, a side swing mechanism and a rotating mechanism. Rotatable shafts are arranged on left and right sides of the elevator body. An inner hinge and an outer hinge are respectively provided on the two rotatable shafts. An inner hinge driving structure includes an inner hinge opening and closing cylinder and an opening and closing cylinder fixing seat. An outer hinge driving structure includes an outer hinge locking cylinder and a locking cylinder fixing seat. The opening and closing cylinder fixing seat and the locking cylinder fixing seat are arranged on a rear side wall of the elevator body. A cylinder barrel part of the inner hinge opening and closing cylinder is hinged on the opening and closing cylinder fixing seat, and a piston part of the inner hinge opening and closing cylinder is hinged to the inner hinge.

Abstract: An image processing method includes: acquiring a standard image matched with an image processing model; determining light field variation parameters of a use environment of the image processing model based on the standard image; determining an image training sample set based on the light field variation parameters, the image training sample set including images of different light field backgrounds; training a correction network of the image processing model through the image training sample set to obtain a model updating parameter adapted to the correction network, and generating a trained image processing model based on the model updating parameter; acquiring a single image in the use environment, and performing prediction processing on the single image through the trained image processing model to obtain light field information corresponding to the single image; and correcting a light field of the single image based on the light field information.

Abstract: This application provides an artificial intelligence-based pathological image processing method performed by an electronic device. The method includes: determining a seed pixel of an immune cell region from a pathological image; obtaining a seed pixel mask image corresponding to the seed pixel of the immune cell region from the pathological image based on the seed pixel of the immune cell region; segmenting an epithelial cell region in the pathological image, to obtain an epithelial cell mask image of the pathological image; fusing the seed pixel mask image and the epithelial cell mask image of the pathological image, to obtain an effective seed pixel mask image corresponding to the immune cell region in the pathological image; and determining a ratio value of the immune cell region in the pathological image based on the effective seed pixel mask image.

Abstract: An image processing method performed by a terminal is provided. A molybdenum target image is obtained, and a plurality of candidate regions are extracted from the molybdenum target image. In the molybdenum target image, a target region is marked in the plurality of candidate regions by using a neural network model obtained by deep learning training, a probability that a lump comprised in the target region is a target lump being greater than a first threshold, a probability that the target lump is a malignant tumor being greater than a second threshold, and the neural network model being used for indicating a mapping relationship between a candidate region and a probability that a lump comprised in the candidate region is the target lump.

Abstract: Embodiments of this application disclose a method and apparatus for training an image region segmentation model, and an image region segmentation method and apparatus. The method includes acquiring a sample image set, and each image of the sample image set having first annotation information; generating graph structure data corresponding to a sample image in the sample image set, the graph structure data comprising multiple nodes, and each node comprising at least one pixel in the sample image; determining second annotation information of each node according to the graph structure data and the first annotation information corresponding to the sample image by using a graph convolutional network model, a granularity of the second annotation information being smaller than a granularity of the first annotation information, the graph convolutional network model being a part of an image segmentation model; and training the image segmentation model according to the second annotation information.

Abstract: The disclosure provides an image processing method and apparatus. The method may include acquiring a target image having at least two division regions. The division region includes an annotation point, and the annotation point is corresponding to a target segmentation object. The method may further include performing iterative expansion on the annotation points to obtain an iterative expansion region. The method may further include determining a first loss result between the iterative expansion region and the division region and a second loss result between the iterative expansion region and the annotation point. The method may further include, when the first loss result and the second loss result meeting a termination condition, stopping the iterative expansion to obtain a resulting iterative expansion region. The method may further include determining the resulting iterative expansion region as a region where the target segmentation object is located in the target image.

Abstract: A hydraulic rotary side-swing elevator includes an elevator body, a side swing mechanism and a rotating mechanism. Rotatable shafts are arranged on left and right sides of the elevator body. An inner hinge and an outer hinge are respectively provided on the two rotatable shafts. An inner hinge driving structure includes an inner hinge opening and closing cylinder and an opening and closing cylinder fixing seat. An outer hinge driving structure includes an outer hinge locking cylinder and a locking cylinder fixing seat. The opening and closing cylinder fixing seat and the locking cylinder fixing seat are arranged on a rear side wall of the elevator body. A cylinder barrel part of the inner hinge opening and closing cylinder is hinged on the opening and closing cylinder fixing seat, and a piston part of the inner hinge opening and closing cylinder is hinged to the inner hinge.

Abstract: The present disclosure relates to a method for training a neural network model performed at an electronic device. The method includes: performing initial training by using a first training sample set to obtain an initial neural network model; performing a prediction on a second training sample set by using the initial neural network model to obtain a prediction result of each of training samples in the second training sample set; determining a plurality of preferred samples from the second training sample set based on the prediction results; adding the plurality of preferred samples that are annotated to the first training sample set to obtain an expanded first training sample set; updating training of the initial neural network model by using the expanded first training sample set to obtain an updated neural network model until a training ending condition is satisfied.

Abstract: A medical image processing method includes: obtaining a biological tissue image including a biological tissue, recognizing, in the biological tissue image, a first region of a lesion object in the biological tissue; recognizing a lesion attribute matching the lesion object; dividing an image region of the biological tissue in the biological tissue image into a plurality of quadrant regions; obtaining target quadrant position information of a quadrant region in which the first region is located; and generating medical service data according to the target quadrant position information and the lesion attribute.