Today, artificial neural networks are trained on large sets of manually tagged images. Generally, for better training, the training data should be as large as possible. Unfortunately, manually tagging images is time consuming and susceptible to error, making it difficult to produce the large sets of tagged data used to train artificial neural networks. To address this problem, the inventors have developed a smart tagging utility that uses a feature extraction unit and a fast-learning classifier to learn tags and tag images automatically, reducing the time to tag large sets of data. The feature extraction unit and fast-learning classifiers can be implemented as artificial neural networks that associate a label with features extracted from an image and tag similar features from the image or other images with the same label. Moreover, the smart tagging system can learn from user adjustment to its proposed tagging. This reduces tagging time and errors.

In a method and system for identifying objects in an image, an image and training data are received. The training data identifies a pixel associated with an object of a particular type in the image. A plurality of filtered versions of the image are developed. The training data and the plurality of filtered versions of the image are processed to develop a trained model for classifying pixels associated with objects of the particular type. The trained model is applied to the image to identify pixels associated a plurality of objects of the particular type in the image. Additional image processing steps are developed to further refine the identified pixels for better fitting of the contour of the objects with their edges.

A computer-implemented tool for automated classification and interpretation of documents, such as life science documents supporting clinical trials, is configured to perform a combination of raw text, document construct, and image analyses to enhance classification accuracy by enabling a more comprehensive machine-based understanding of document content. The combination of analyses provides context for classification by leveraging relative spatial relationships among text and image elements, identifying characteristics and formatting of elements, and extracting additional metadata from the documents as compared to conventional automated classification tools, wherein natural language processing (NLP) is applied to associate text with tokens, and relevant differences and similarities between protocols are identified.

A machine learning (ML) task system trains a neural network model that learns a compressed representation of acquired data and performs a ML task using the compressed representation. The neural network model is trained to generate a compressed representation that balances the objectives of achieving a target codelength and achieving a high accuracy of the output of the performed ML task. During deployment, an encoder portion and a task portion of the neural network model are separately deployed. A first system acquires data, applies the encoder portion to generate a compressed representation, performs an encoding process to generate compressed codes, and transmits the compressed codes. A second system regenerates the compressed representation from the compressed codes and applies the task model to determine the output of a ML task.

A method for receipt processing using optical character recognition (OCR). The method includes detecting, within a receipt image, a pre-determined characteristic of the receipt, obtaining an OCR receipt template based on the pre-determined characteristic of the receipt, extracting, based on at least the OCR receipt template, a vendor attribute from the receipt image corresponding to a pre-defined field of the receipt, verifying, based on matching the vendor attribute to a vendor associated with the OCR receipt template, that the receipt is generated by the vendor and that the OCR receipt template is applicable to perform the OCR of the receipt image, and generating, in response to the verifying, a textual content of the receipt by processing the receipt image based at least on the OCR receipt template.

This disclosure relates to system and method for optical character recognition. In one embodiment, the method comprises providing an image data to a plurality of customized machine learning algorithms or various customized neural networks, configured to recognize a set of pre-defined characters. The method comprises presenting one or more suggestions for the character to the user in response to negative character recognition, and training a customized machine learning algorithm corresponding to the character if one of the suggestions is identified by the user. If the suggestions are rejected by the user, the method comprises prompting the user to identify the character and determining presence of the character in the set of pre-defined characters. The method further comprises training a customized machine learning algorithm corresponding to the character if the character is present, or dynamically creating a customized machine learning algorithm corresponding to the character if the character is not present.

A display method includes acquiring captured image data generated by capturing an image of a second area containing a first area, the captured image data containing first partial data representing an image of the first area and second partial data representing an image of the second area, recognizing an object located in the first area by using the first partial data with no use of the second partial data, displaying a first image that underwent a first process according to the position of the object when the object is recognized as a first pointing element and when the object moves from the first area onto a display surface, and displaying a second image that underwent a second process according to the position of the object when the object is recognized as a second pointing element and when the object moves from the first area onto the display surface.

A computer vision (CV) training system, includes: a supervised learning system to estimate a supervision output from one or more input images according to a target CV application, and to determine a supervised loss according to the supervision output and a ground-truth of the supervision output; an unsupervised learning system to determine an unsupervised loss according to the supervision output and the one or more input images; a weakly supervised learning system to determine a weakly supervised loss according to the supervision output and a weak label corresponding to the one or more input images; and a joint optimizer to concurrently optimize the supervised loss, the unsupervised loss, and the weakly supervised loss.

A computer-implemented method, a computing system, and a computer program product, for automatically labeling an amount of unlabeled data for training one or more classifiers of a machine learning system. A method includes iteratively processing unlabeled data items. Receiving an unlabeled data item into each autoencoder in an autoencoder architecture. Each autoencoder processing with a lowest loss of information the unlabeled data item that is likely associated with a label associated with the autoencoder, while processing with a higher loss of information the unlabeled data item that is likely not associated with the label. Predicting, based on loss of information, a probability distribution for the unlabeled data item. Automatically associating the label to the unlabeled data item, based on the label being associated with a highest probability in a peaking probability distribution associated with the unlabeled data item. The autoencoder architecture can include a cloud computing network architecture.

An improved method for verifying whether a character-recognition technology has correctly identified which characters are represented by character images involves displaying the uncertain character images in place of their respective hypothesis characters in a document being read a verifier. The verifier may mark incorrectly spelled words containing the uncertain character images. Based on the markings, a system adjusts a confidence level associated with the hypothesis about the uncertain character in order to obtain a confirmed hypothesis linked to the uncertain character.