This disclosure describes systems, methods, and devices for artificial intelligence-based grading and feedback of digitally entered handwritten characters into a device. A method may include converting, using a first device, a computer-readable document with questions into a digital worksheet comprising teacher layers and student layers; detecting, using a first machine learning model trained to categorize questions and generate answer zones for the questions, answer zones; generating first updated student layers comprising the questions and the answer zones; receiving second updated student layers comprising the first updated student layers and respective answers digitally handwritten into the answer zones; generating, using a second machine learning model, clusters of the respective answers based on hand stroke similarities in the respective answers and based on content similarities in the respective answers; and presenting, using the first device and the teacher layers, the respective answers based on the clusters.

A manual providing apparatus includes: circuitry; and a memory storing computer-executable instructions that cause the circuitry to execute: acquiring image data in a manual; performing a first detection of detecting a position of a target component included in the image data in the manual; performing a second detection of detecting a scope of a surrounding component included in the image data in the manual; acquiring relative position information representing a relative position between the target component and the surrounding component, based on the detected position of the target component and the detected scope of the surrounding component; and outputting work procedure information in which the relative position information is added to text data in the manual.

Systems and methods of verifying pairings of price tag labels and products at a product storage facility include an image capture device that moves about and captures images of product storage structures at the product storage facility, and a database that stores inventory data associated with the products stocked at the product storage facility. A computing device processes the images to extract one or more characters from the price tag labels and products detected in the images, correlates the extracted characters to the inventory data to identify a product code and a product name that matches the characters extracted from the price tag labels and the products. If the characters extracted from the price tag and from the on-shelf product to which the price tag label is allocated match, the database is updated to reflect that the price tag label is properly allocated to the appropriate on-shelf product.

A method includes receiving a binary annotation of source text; performing a close operation on the binary annotation to generate a closed annotation using an initial kernel size; defining one or more contours in the closed annotation using one or more bounding boxes, respectively; determining a subset of the one or more contours for which a percentage of area occupied by text within a corresponding bounding box exceeds a threshold; and generating a final annotation of the source text based on the subset of the one or more contours.

An online handwriting document layout analysis system includes a preprocess unit serving to receive a document composed of a plurality of strokes and to generate an undirected graph including a plurality of nodes and a plurality of edges for representing relations between different strokes. A bidirectional recursive neural network unit for initializing a feature vector of each of the nodes and initializing a feature vector of each of edges. A graphic neural network unit serves to update the feature vectors of the nodes and the edges for obtaining updated feature vectors. A fully connected neural network unit serves for performing a coarse-grained object classifying and a fine-grained object classifying for each of the nodes and the edges based on the updated feature vectors. A document restoration unit serves for restoring a tree structure of the document.

An image processing system includes a target image acquiring unit, an object detecting unit, an area extracting unit, and a process executing unit. The target image acquiring unit is configured to acquire as a target image a document image of a document. The object detecting unit is configured to detect an area specifying object additionally written to the document by handwriting in the document image using object detection with a learner for which machine learning has been performed. The area extracting unit is configured to extract a free shape area specified by the detected area specifying object. The process executing unit is configured to execute a predetermined process for the free shape area extracted in the target image.

An electronic device includes: a memory storing one or more instructions; when executed by at least one processor, cause the electronic device to: record an external voice for a certain period of time, recognize handwritten characters input based on an external input signal, generate handwritten text by converting the handwritten characters to text, determine whether there is a missing symbol based on the handwritten text, obtain, based on determining that there is the missing symbol, a voice segment from the recorded external voice recorded for the certain period of time, the voice segment corresponding to a time at which the missing symbol is input, generate target text by converting a voice included in the voice segment to text, determine recommended text corresponding to a location of the missing symbol based on the target text and the handwritten text, and change the missing symbol to the recommended text.

This application provides a method of detecting printing defects. The method includes obtaining a first image of each character in a reference image. A third image of each character is obtained based on the first image of each character, a fourth image of each character is obtained based on a second image of each character obtained from an image to be detected. Once a fifth image of each character is obtained based on the third image of each character, a sixth image of each character is obtained according to the fourth image and the fifth image of each character, a detection result of each character in the image to be detected is determined according to the fifth image and the sixth image of the each character.

A processor of a wearable device is provided. The processor may identify an external object included in a displaying area of a display by using a camera. The processor may display a first visual object representing an application corresponding to the external object based on the location of the external object in the displaying area. The processor may obtain at least one of first information associated with a path of the external object moved on the plane, or second information including at least one stroke drawn by the external object, while the first visual object is displayed. The processor may display a second visual object for executing the application by using information, selected among the first information or the second information, based on the location, according to the attribute assigned to the external object.

A method of processing an image, performed by a computing apparatus including a processor, according to some exemplary embodiments of the present disclosure, may include: obtaining an image; obtaining, from the image, analysis information corresponding to an object included in the image by using an object analysis model; and obtaining, from the analysis information corresponding to the object, a character contained in the object by using an OCR model.