Techniques for classifying biomedical image data using a graph neural network are disclosed. In one particular embodiment, the techniques may be realized as a method for classifying biomedical image data comprising generating an annotated representation of biomedical image data; identifying a plurality of pixel clusters based on the biomedical image data; constructing a graph based on the plurality of pixel clusters; determining at least one biomedical feature for at least one node of the graph based on the annotated representation of the biomedical image data; and processing a graph neural network to classify the biomedical image data based on the at least one biomedical feature.

Proposed is a technique that can detect a random noise component at high accuracy without measurement pattern limitation and enables edge roughness measurement at higher accuracy. According to this disclosure, pattern matching and edge position correction are performed with respect to each of the left edge and the right edge of a line pattern in an obtained line pattern image, and an image with no roughness is generated. A PSD value is measured from the image, and the average PSD value of all the frequencies is determined as a random noise component, so that the random noise component can be detected at high accuracy. Further, the PSD value (random noise component) is subtracted from the PSD value of an original image, thereby measuring edge roughness at high accuracy.

A non-transitory computer-readable storage medium, a trace detection device, and a device and method for training a trace detection model are described, relating to the technical field of machine learning. The training method comprises obtaining a sample image and a sample tag of the sample image; performing line segment detection on the sample image, and obtaining a line segment edge feature of the sample image; generating a training feature according to the line segment edge feature; and training a classification model according to the sample tag and the training feature to obtain a trace detection model (S140).

Screen response validation of robot execution for robotic process automation (RPA) is disclosed. Whether text, screen changes, images, and/or other expected visual actions occur in an application executing on a computing system that an RPA robot is interacting with may be recognized. Where the robot has been typing may be determined and the physical position on the screen based on the current resolution of where one or more characters, images, windows, etc. appeared may be provided. The physical position of these elements, or the lack thereof, may allow determination of which field(s) the robot is typing in and what the associated application is for the purpose of validation that the application and computing system are responding as intended. When the expected screen changes do not occur, the robot can stop and throw an exception, go back and attempt the intended interaction again, restart the workflow, or take another suitable action.

Provided is a bending angle calculation method capable of saving a worker time and effort. The bending angle calculation method includes: an image capturing step (S1) of capturing an image of a pipe in which a first pipe and a second pipe are joined together by a joint; a derivation step (S3) of deriving, from the image, a laying direction straight line corresponding to a laying direction of the pipe; and a calculation step (S4) of calculating, as a bending angle at the joint, a crossing angle between a laying direction straight line of the first pipe and a laying direction straight line of the second pipe.

According to one embodiment, a reading system includes an extractor, a determiner, and a reader. The extractor extracts a candidate image from an input image. The candidate image is of a portion of the input image in which a segment display is imaged. The determiner calculates an angle with respect to a reference line of each of a plurality of straight lines detected from the candidate image, and determines whether or not the candidate image is an image of a segment display based on a distribution indicating a relationship between the angle and a number of the straight lines. The reader reads a numerical value displayed in a segment display from the candidate image determined to be an image of a segment display.

Various embodiments provide a polygonal region detection method and apparatus, a computer readable storage medium and an electronic device. In those embodiments, a to-be-detected image can be obtained. A plurality of line segments in the image can be calculated based on a line detection algorithm. The plurality of line segments meeting a merging condition can be merged into a line segment. Crosspoints of the pairwise merged line segments can be calculated according to the merged line segments in the image. A polygonal region can be generated with the crosspoints as vertexes of the polygonal region in the image.

According to one embodiment, a reading system includes an extractor, a determiner, and a reader. The extractor extracts a candidate image from an input image. The candidate image is of a portion of the input image in which a segment display is imaged. The determiner calculates an angle with respect to a reference line of each of a plurality of straight lines detected from the candidate image, and determines whether or not the candidate image is an image of a segment display based on a distribution indicating a relationship between the angle and a number of the straight lines. The reader reads a numerical value displayed in a segment display from the candidate image determined to be an image of a segment display.

A computer implemented method comprises receiving one or more images of a computing environment comprising a plurality of interconnected components, analyzing the or each received image to identify each component shown in the image(s) and the connection(s) of each identified component, by identifying a set of attributes for each component from the image(s) and matching the identified attributes to attributes of known components stored in a database, obtaining a specification for each identified component, and generating a document comprising each identified component, its respective specification and the connection(s) of each identified component.

The present disclosure provides a method for automatically identifying a ring joint of a shield tunnel based on a lining structure. The method includes the following steps: S1: acquiring a three-dimensional (3D) point cloud of a shield tunnel through a mobile scanning system: S2: generating an orthographic projection image of an inner wall of the tunnel: S3: identifving a feature of a bolt hole; S4: extracting a longitudinal joint of the shield tunnel: S5: generating a prior structural template ring; and S6: extracting a transverse joint of the shield tunnel. The present disclosure has the following advantages. Starting from the features of the lining structure of the shield tunnel, the present disclosure selects a bolt hole with a strong structural feature, takes the structural feature of the bolt hole as an identification feature, and indirectly extracts joint information of straight and staggered joints tunnel.