Devices and techniques are generally described for per-query prediction of shard relevance for search. In some examples, a search system may receive a first search query. A first score may be determined for a first database partition, the first score indicating a relevancy of the first search query to first data stored by the first database partition. Similarly, a second score may be determined for a second database partition, the second score indicating a relevancy of the first search query to second data stored by the second database partition. A determination may be made that the first search query is related to the first data stored by the first database partition. A determination may be made, based at least in part on the second score, that the first search query is unrelated to the second data stored by the second database partition.

System and method to recognise characters from an image are disclosed. The method includes receiving the at least one image, pre-processing the at least one image, extracting a plurality of characters from the corresponding at least one image, extracting at least one structure from the corresponding at least one image upon applying an edge detection technique to extract a structure, identifying a template based on extracted structure, subjecting the plurality of characters into a plurality of ensemble AI models to extract one of a plurality of texts, a plurality of non-textual data and a combination thereof, comparing a plurality of extracted plurality of texts, a plurality of non-textual data, or a combination thereof from the corresponding plurality of ensemble AI models with each other, generating a confidence score and validating one of the plurality of accurate texts, the plurality of accurate non-textual data, or a combination thereof.

Computer-implemented methods, systems and devices having segmented numeral character recognition. In an embodiment, users may take digital pictures of a seven-segment display on a sensor device. For example, a user at a remote location may use a digital camera to capture a digital image of a seven-segment display on a sensor device. Captured images of a seven-segment display may then be sent or uploaded over a network to a remote health management system. The health care management system includes a reader that processes the received images to determine sensor readings representative of the values output on the seven-segment displays of the remote sensor devices. Machine learning and OCR are used to identify numeric characters in images associated with seven-segment displays. In this way, a remote heath management system can obtain sensor readings from remote locations when users only have access to sensor devices with seven-segment displays.

The present disclosure discloses a method for training an adversarial network model, a method for building a character library, an electronic device and a storage medium, which relate to a field of artificial intelligence, in particular to a field of computer vision and deep learning technologies, and are applicable in a scene of image processing and image recognition. The method for training includes: generating a new character by using the generation model based on a stroke character sample and a line character sample; discriminating a reality of the generated new character by using the discrimination model; calculating a basic loss based on the new character and a discrimination result; calculating a track consistency loss based on a track consistency between the line character sample and the new character; and adjusting a parameter of the generation model according to the basic loss and the track consistency loss.

A method for testing medical data is provided. Each medical datum includes a plurality of information units and a plurality of separators, and the method includes the following steps: a. matching the medical data against a standard library including a plurality of patterns, a matching expression being:

[\s\S][number/sequence/relation]&[\b|\B] (S101); and b. determining, based on a matching result of the step a, whether the medical datum is qualified (S102). A standardized standard library is first established, a matching result is obtained by matching the medical datum and the standard library for a non-initial boundary, an initial boundary, an information quantity, information sequences, a semantic relationship quantity, a character boundary, and a non-character boundary, and whether the medical datum meets a requirement is further determined according to the matching result.

System and method to recognise characters from an image are disclosed. The method includes receiving the at least one image, pre-processing the at least one image, extracting a plurality of characters from the corresponding at least one image, extracting at least one structure from the corresponding at least one image upon applying an edge detection technique to extract a structure, identifying a template based on extracted structure, subjecting the plurality of characters into a plurality of ensemble AI models to extract one of a plurality of texts, a plurality of non-textual data and a combination thereof, comparing a plurality of extracted plurality of texts, a plurality of non-textual data, or a combination thereof from the corresponding plurality of ensemble AI models with each other, generating a confidence score and validating one of the plurality of accurate texts, the plurality of accurate non-textual data, or a combination thereof.

Techniques for automatically extracting data from documents using multiple deep learning models are provided. According to one set of embodiments, a computer system can receive a document in an electronic format and can segment, using an image segmentation deep learning model, the document into a plurality of segments, where each segment corresponds to a visually discrete portion of the document and is classified as being one of a plurality of types. The computer system can then, for each segment in the plurality of segments, retrieve text in the segment using optical character recognition (OCR) and extract data in the segment from the retrieved text using a named entity recognition (NER) deep learning model, where the retrieving and the extracting are performed in a manner that takes into account the segment's type.

Aspects of the present disclosure relate to audio/video (A/V) stream functionality verification. A stream segment of a video feed prior to transmission over a network as captured by a transmitting device within a web-based conference can be stored. A stream segment of the video feed after transmission over the network as received by a receiving device within the web-based conference can be stored. The stream segment of the video feed prior to transmission over the network can be compared with the stream segment of the video feed after transmission over the network to determine a video feed quality.

A vision-based product inspection system captures multiple images of each of multiple individual instances of a product as each instance passes through various phases of a production process. The system includes multiple cameras with each camera situated at a known location along a moving conveyor, conveyor belt, production line, or assembly line that moves instances of the product through various phases of the production process. Each camera can be associated with a known location along the conveyor and each image can be associated with a value representing the position of the conveyor as it moves product. Based on each camera's location and the values representing the conveyor's position, a sequence of images can be accumulated representing the progression of any single instance of a product as it moves through the production process. Automated quality control inspection can be performed by comparing or analyzing images in the sequence.

A license plate recognition process is described that includes automated image analysis integrated with human review and a definition of a plate grammar of license plate visual elements that are required to prepare a database search index that enables selection of a particular license plate, from a known population of all possible license plates issued for each and every jurisdiction of interest. The plate grammar includes visual elements that may be searched by both automated image analysis and by a human reviewer. The process includes creation of a question that guides a human reviewer into providing required information regarding a selected plate grammar element that is not identified by computerized image analysis. Integration of a manual review process within an automated process as opposed to manual review solely at a failed completion of the automated process is included. Methods are presented for creation of queries for manual review that take into account the skill and availability of the manual reviewer as well as optimizing for a shortest path to completion of the recognition of the license plate through the database search index.