Embodiments of the present disclosure pertain to extracting data corresponding to particular data types using machine learning algorithms. In one embodiment, a method includes receiving an image in a backend system, sending the image to an optical character recognition (OCR) component, and in accordance therewith, receiving a plurality of characters recognized in the image. The character set is matched against known values to generate candidate character strings. The character set is processed by one or more machine learning algorithms to produce features. For each candidate character string, the features are then processed by a random forest model to determine a final character string.

An image reader includes a document reading unit, and a control unit that functions as an individual image cutting section, character string detection section, mismatch detection section, judgment section, and correction section. The individual image cutting section cuts out individual images from image data obtained through reading by the document reading unit. The character string detection section detects character strings present on the individual images. The mismatch detection section detects, for the character strings detected by the character string detection section, a mismatching portion by making comparison between the individual images with considering character strings having contents identical or similar to each other as same information. The judgment section judges for the mismatching portions whether a ratio of majority characters reaches a predefined ratio. Upon judging that the ratio of the majority characters has reached the predefined ratio, the correction section replaces a minority character with the majority character.

The positioning of elements of a broken word can be corrected by receiving an optical character recognition (OCR) conversion of a printed publication and identifying multiple parts of the broken word from the OCR conversion to output in a graphical user interface (GUI). The multiple parts can be placed in the GUI using original positioning data for the printed publication. A user can make a selection in the GUI indicating that multiple parts from the OCR conversion are of the broken word and can automatically adjust bounds of the multiple parts to form a corrected word.

An object recognition device according to an embodiment includes a camera that captures an image of an imaging area. A storage device stores, for each of a plurality of registered objects, dictionary feature information for identifying the corresponding object and dictionary boundary information for identifying an actual boundary area of the corresponding object. A processor receives the captured image from the camera, and determines an object area in the captured image. The processor extracts feature information from the object area, and, based on the extracted feature information compared to the dictionary feature information, identifies each object included in the object area. The processor also extracts boundary information corresponding to each identified object included in the object area, and, based on the extracted boundary information compared to the dictionary boundary information with respect to each identified object, determines an overlap state of each identified object in the object area.

Systems and techniques are disclosed for labeling objects within an image. The objects may be labeled by selecting an option from a plurality of options such that each option is a potential label for the object. An option may have an option score associated with. Additionally, a relation score may be calculated for a first option and a second option corresponding to a second object in an image. The relation score may be based on a frequency, probability, or observance corresponding to the co-occurrence of text associated with the first option and the second option in a text corpus such as the World Wide Web. An option may be selected as a label for an object based on a global score calculated based at least on an option score and relation score associated with the option.

The present teaching relates to method, system, and programming for responding to an image related query. Information related to each of a plurality of images is received, wherein the information represents concepts co-existing in the image. Visual semantics for each of the plurality of images are created based on the information related thereto. Representations of scenes of the plurality of images are obtained via machine learning, based on the visual semantics of the plurality of images, wherein the representations capture concepts associated with the scenes.

According to an embodiment, a recognition device includes a detector, a recognizer, and a matcher. The detector is configured to detect a character candidate from an input image. The recognizer is configured to generate recognition candidate from the character candidate. The matcher is configured to match the recognition candidate with a knowledge dictionary and contains modeled character strings to be recognized, and generate a matching result obtained by matching a character string presumed to be included in the input image with the dictionary. Any one of a real character code that represents a character and a virtual character code that specifies a command is assigned to an edge. The matcher gives, when shifting a state of the dictionary in accordance with an edge to which the virtual character code is assigned, a command specified by the virtual character code assigned to the edge to a command processor.

A method and an apparatus for recognizing characters using an image are provided. A camera is activated according to a character recognition request and a preview mode is set for displaying an image photographed through the camera in real time. An auto focus of the camera is controlled and an image having a predetermined level of clarity is obtained for character recognition from the images obtained in the preview mode. The image for character recognition is character-recognition-processed so as to extract recognition result data. A final recognition character row is drawn that excludes non-character data from the recognition result data. A first word is combined including at least one character of the final recognition character row and a predetermined maximum number of characters. A dictionary database that stores dictionary information on various languages using the first word is searched, so as to provide the user with the corresponding word.

Techniques are disclosed for generating a sequence of symbols based on input data for a neuro-linguistic model. The model may be used by a behavior recognition system to analyze the input data. A mapper component of a neuro-linguistic module in the behavior recognition system receives one or more normalized vectors generated from the input data. The mapper component generates one or more clusters based on a statistical distribution of the normalized vectors. The mapper component evaluates statistics and identifies statistically relevant clusters. The mapper component assigns a distinct symbol to each of the identified clusters.

The present disclosure describes devices and methods of providing a technology environment for analyzing unstructured data to generate structured data. A set of electronic documents, each electronic document associated with a type of product, may be accessed. A data instance for each of the documents may be generated. The data instance may include a plurality of data fields that are based on the type of product. The electronic documents may be analyzed to identify values for each of the plurality of data fields. Analyzing the electronic documents may comprise applying a respective character recognition algorithm to respective electronic documents, and assigning a confidence factor to each of the values. The data instances comprising the values for each of the plurality of data fields may be stored in a second database.