Optical character recognition (OCR) based systems and methods for extracting and automatically evaluating contextual and identification information and associated metadata from an image utilizing enhanced image processing techniques and image segmentation. A unique, comprehensive integration with an account provider system and other third party systems may be utilized to automate the execution of an action associated with an online account. The system may evaluate text extracted from a captured image utilizing machine learning processing to classify an image type for the captured image, and select an optical character recognition model based on the classified image type. They system may compare a data value extracted from the recognized text for a particular data type with an associated online account data value for the particular data type to evaluate whether to automatically execute an action associated with the online account linked to the image based on the data value comparison.

Some embodiments of the disclosure provide a diabetic retinopathy recognition system (S) based on fundus image. According to an embodiment, the system includes an image acquisition apparatus (1) configured to collect fundus images. The fundus images include target fundus images and reference fundus images taken from a person. The system further includes an automatic recognition apparatus (2) configured to process the fundus images from the image acquisition apparatus by using a deep learning method. The automatic recognition apparatus automatically determines whether a fundus image has a lesion and outputs the diagnostic result. According to another embodiment, the diabetic retinopathy recognition system (S) utilizes a deep learning method to automatically determine the fundus images and output the diagnostic result.

The Parkinson's disease information providing apparatus using a neuromelanin image according to an aspect of the present disclosure includes an image receiving unit which acquires an MRI image obtained by capturing a brain of a patient; an image preprocessing unit which preprocesses the acquired MRI image to observe the neuromelanin region used as an image bio marker of the Parkinson's disease; an image processing unit which analyzes the preprocessed MRI image to classify a first image including the neuromelanin region and detects the neuromelanin region from the classified first image; and an image analyzing unit which diagnoses whether the patient has the Parkinson's disease by analyzing whether the detected the neuromelanin region is normal.

In various examples, the present disclosure relates to using temporal filters for automated real-time classification. The technology described herein improves the performance of a multiclass classifier that may be used to classify a temporal sequence of input signals—such as input signals representative of video frames. A performance improvement may be achieved, at least in part, by applying a temporal filter to an output of the multiclass classifier. For example, the temporal filter may leverage classifications associated with preceding input signals to improve the final classification given to a subsequent signal. In some embodiments, the temporal filter may also use data from a confusion matrix to correct for the probable occurrence of certain types of classification errors. The temporal filter may be a linear filter, a nonlinear filter, an adaptive filter, and/or a statistical filter.

A remote image interpretation management apparatus includes a hardware processor. The hardware processor obtains, from a storage storing evaluation values about quality of image interpretation reports by evaluator and by image interpretation facility and/or image interpretation doctor, evaluation values of evaluators who have evaluated a predetermined number of image interpretation reports or more. Based on the obtained evaluation values of the evaluators, for each of the evaluators, the hardware processor calculates a statistic of the evaluation values of the evaluator and normalizes the evaluation values with the calculated statistic, thereby obtaining normalized evaluation values for the respective evaluators. For each image interpretation facility and/or each image interpretation doctor, the hardware processor calculates a mean value based on the normalized evaluation values of the respective evaluators.

In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

The present disclosure provides a method in a data processing system that includes at least one processor and at least one memory. The at least one memory includes instructions executed by the at least one processor to implement a driving encounter recognition system. The method includes receiving information, from one or more sensors coupled to a first vehicle, determining first trajectory information associated with the first vehicle and second trajectory information associated with a second vehicle, extracting a feature vector, providing the feature vector to a trained classifier, the classifier trained using unsupervised learning based on a plurality of feature vectors, and receiving, from the trained classifier, a classification of the current driving encounter in order to facilitate the first vehicle to perform a maneuver based on the current driving encounter.

Various techniques employ digital fingerprints, alone or in conjunction with other data related to a physical object, to characterize the physical object in a way allows an identity of the physical object to be confirmed or authenticated within a threshold of confidence when the physical object is encountered at a subsequently after a digital fingerprint thereof was previously ingested, without capture, retention and/or reference to certain types or categories of characterizing information which may be of a sensitive nature and/or subject to misuse or abuse. Such digital fingerprints may include feature vectors and non-positional characterization values, while advantageously omitting positional information that would otherwise allow recreation of a recognizable image of the physical object or an image from which one or more immutable characteristics or traits (e.g., physical appearance, color, race, ethnicity, gender, age) of the physical object may be discerned, and hence denominated as irreversible digital fingerprints.

A method, non-transitory computer readable medium and apparatus for performing a person re-identification using an overhead view image are disclosed. For example, the method includes receiving a plurality of overhead view images, detecting a target person in one or more of the plurality of overhead view images, creating a probe image of the target person, receiving a selection of the probe image containing the target person, selecting one or more of the plurality of processed images that has a similar distortion profile as a distortion profile of the probe image based on a radial distance of the target person from a center of a respective overhead view image of the plurality overhead view images used to generate the probe image and performing the person-re-identification of the target person using the one or more of the plurality of processed images that are selected.

Systems and methods are described for Distification of 3D imagery. A computing device may obtain a three dimensional (3D) image that includes rules defining a 3D point cloud used to generate a two dimensional (2D) image matrix. The 2D image matrix may include 2D matrix point(s) mapped to the 3D image, where each 2D matrix point can be associated with a horizontal coordinate and a vertical coordinate. The computing device can generate an output feature vector that includes, for at least one of the 2D matrix points, the horizontal coordinate and the vertical coordinate of the 2D matrix point, and a depth coordinate of a 3D point in the 3D point cloud of the 3D image. The 3D point can have a nearest horizontal and vertical coordinate pair that corresponds to the horizontal and vertical coordinates of the at least one 2D matrix point.