Abstract: Embodiments of this application provide a defect detection method and apparatus. The method includes: obtaining an image for inspection; performing anomaly detection on the image for inspection to obtain an anomaly region image corresponding to the image for inspection; and performing defect classification on the anomaly region image to obtain defect detection information of the image for inspection. The defect detection method of the embodiments of this application is divided into two steps of anomaly detection and defect classification. Anomaly detection is performed on the image for inspection first, and then defect classification needs to be performed only on an anomaly region, reducing the workload of defect classification, thereby improving the efficiency of defect detection.

Abstract: The present application relates to an image processing method and system. The method may include: acquiring a sequence of input images containing a target object; and performing multi-resolution fusion on the sequence of input images to generate a single fused image, where pixels of the fused image may include a pixel at a corresponding position of an input image in the sequence of input images, and each pixel of the fused image containing the target object may include a pixel at a corresponding position of an input image in the sequence of input images in which part of the target object is focused.

Abstract: A tab bending detection method and apparatus, an electronic device, and a storage medium are provided. The method includes: performing skeleton extraction on a sectional image of multiple layers of tabs to obtain a skeleton image of the multiple layers of tabs; merging damaged connected components in the skeleton image to obtain a merged connected component, where the damaged connected components are connected components on which breaking occurs in a same tab section; calculating a target number of the multiple layers of tabs based on the merged connected component and an undamaged connected component; and detecting, based on the target number and a preset number, whether any tab in the multiple layers of tabs is in a bending state. The damaged connected components are merged to obtain the merged connected component.

Abstract: An image processing method includes searching for a next pixel of a skeleton branch of a target image based on an already-searched current pixel of the skeleton branch, and determining whether the skeleton branch is a burr branch based on a number of already-searched pixels of the skeleton branch.

Abstract: An image processing method includes searching for a next pixel of a skeleton branch of a target image based on an already-searched current pixel of the skeleton branch, and determining whether the skeleton branch is a burr branch based on a number of already-searched pixels of the skeleton branch.

Abstract: Embodiments of this application provide a defect detection method and apparatus. The method includes: obtaining an image for inspection; performing anomaly detection on the image for inspection to obtain an anomaly region image corresponding to the image for inspection; and performing defect classification on the anomaly region image to obtain defect detection information of the image for inspection. The defect detection method of the embodiments of this application is divided into two steps of anomaly detection and defect classification. Anomaly detection is performed on the image for inspection first, and then defect classification needs to be performed only on an anomaly region, reducing the workload of defect classification, thereby improving the efficiency of defect detection.

Abstract: A defect detection method includes: obtaining an average grayscale value of an image; constructing a mapping table, where elements of the mapping table include a mapped value corresponding to each grayscale value within a grayscale value range of the image, a mapped value corresponding to a grayscale value greater than or equal to a reference value is a first value, a mapped value corresponding to a grayscale value less than the reference value is a second value, and the reference value is an absolute value of a difference between the average grayscale value and a preset grayscale value; searching for a mapped value corresponding to a grayscale value of each pixel in the image from the mapping table; segmenting the image to obtain a suspicious defect sub-image based on the mapped value corresponding to the grayscale value of each pixel; and inputting the suspicious defect sub-image into a machine learning model.

Abstract: An image recognition method may include: acquiring a target image, where the target image may include a weld bead region; performing initial segmentation on the target image, to obtain a first recognition result, where the first recognition result may include first recognition information for the weld bead region in the target image; performing feature extraction on the target image, to obtain a region representation; obtaining a context representation based on the first recognition result and the region representation, where the context representation may be used for representing a correlation between each pixel and remaining pixels in the target image; and obtaining a second recognition result based on the context representation, where the second recognition result may include second recognition information for the weld bead region in the target image.

Abstract: The present application discloses a method for inspecting a battery tab, the method including: obtaining a sectional view of a plurality of layers of tabs of a battery; identifying and analyzing the sectional view to obtain a plurality of connected domains, where each connected domain includes one tab or a plurality of tabs that are bonded with each other; determining, based on positions and a number of intersection points of tab bonding in each connected domain, a number of layers of tabs corresponding to the connected domain; calculating a total number of layers of the plurality of layers of tabs in the sectional view based on the number of layers of tabs corresponding to the connected domain; and determining, based on the total number of layers of tabs and a preset real number of layers of tabs, whether the plurality of layers of tabs are folded.

Abstract: The present application discloses a method for inspecting a battery tab, the method including: obtaining a sectional view of a plurality of layers of tabs of a battery; identifying and analyzing the sectional view to obtain a plurality of connected domains, where each connected domain includes one tab or a plurality of tabs that are bonded with each other; determining, based on positions and a number of intersection points of tab bonding in each connected domain, a number of layers of tabs corresponding to the connected domain; calculating a total number of layers of the plurality of layers of tabs in the sectional view based on the number of layers of tabs corresponding to the connected domain; and determining, based on the total number of layers of tabs and a preset real number of layers of tabs, whether the plurality of layers of tabs are folded.

Abstract: The present application relates to an image processing method and system. The method may include: acquiring a sequence of input images containing a target object; and performing multi-resolution fusion on the sequence of input images to generate a single fused image, where pixels of the fused image may include a pixel at a corresponding position of an input image in the sequence of input images, and each pixel of the fused image containing the target object may include a pixel at a corresponding position of an input image in the sequence of input images in which part of the target object is focused.

Abstract: An image recognition method may include: acquiring a target image, where the target image may include a weld bead region; performing initial segmentation on the target image, to obtain a first recognition result, where the first recognition result may include first recognition information for the weld bead region in the target image; performing feature extraction on the target image, to obtain a region representation; obtaining a context representation based on the first recognition result and the region representation, where the context representation may be used for representing a correlation between each pixel and remaining pixels in the target image; and obtaining a second recognition result based on the context representation, where the second recognition result may include second recognition information for the weld bead region in the target image.