Computer-implemented methods for detecting objects within digital image data based on color transitions include: receiving or capturing a digital image depicting an object; sampling color information from a first plurality of pixels of the digital image, wherein each of the first plurality of pixels is located in a background region of the digital image; assigning each pixel a label of either foreground or background using an adaptive label learning process; binarizing the digital image based on the labels assigned to each pixel; detecting contour(s) within the binarized digital image; and defining edge(s) of the object based on the detected contour(s). Corresponding systems and computer program products configured to perform the inventive methods are also described.

Techniques for processing an image of a check captured using a mobile device are provided. The check image is processed to determine whether the check can be deposited at a bank via a mobile deposit process. The system can identify regions of the check—such as the endorsement area—to determine if the check has been properly endorsed. The system can be implemented on a mobile device and/or a server, where the mobile device routes the check image to the server for processing. If the check cannot be deposited, a rejection is forwarded in real time to the mobile device for possible correction.

Methods for identifying and tracking an article are disclosed. During a journey, an article follows a path between an origin and a destination. An image of the article is captured during the journey and a first characteristic vector determined from the image of the article. The first characteristic vector is compared with a set of predetermined characteristic vectors and, based on the comparison, the first article is either associated with an identifier associated with a corresponding one of the predetermined characteristic vectors, or is associated with a new identifier.

A processor may receive an image and determine a number of foreground pixels in the image. The processor may obtain a result of optical character recognition (OCR) processing performed on the image. The processor may identify at least one bounding box surrounding at least one portion of text in the result and overlay the at least one bounding box on the image to form a masked image. The processor may determine a number of foreground pixels in the masked image and a decrease in the number of foreground pixels in the masked image relative to the number of foreground pixels in the image. Based on the decrease, the processor may modify an aspect of the OCR processing for subsequent image processing.

A processor may receive an image and determine a number of foreground pixels in the image. The processor may obtain a result of optical character recognition (OCR) processing performed on the image. The processor may identify at least one bounding box surrounding at least one portion of text in the result and overlay the at least one bounding box on the image to form a masked image. The processor may determine a number of foreground pixels in the masked image and a decrease in the number of foreground pixels in the masked image relative to the number of foreground pixels in the image. Based on the decrease, the processor may modify an aspect of the OCR processing for subsequent image processing.

A method for structural analysis and recognition of a handwritten mathematical formula in a natural scene image, including: transforming a gray matrix of a natural scene image into a local contrast matrix, and performing a binary division to the obtained local contrast matrix using an Otsu method, thereby obtaining a binary matrix; performing a connected domain analysis to the binary matrix, eliminating non-character connected domains to obtain character connected domains; performing a detection of elements of a special structure of a formula to the character connected domains using a correlation coefficient method, and separately annotating all the detected elements of the special structure: dividing rows of the binary matrix by means of horizontal projection; recognizing each character connected domain by means of a convolutional neural network; defining an output sequence, and outputting the results of recognition in a corresponding sequence according to a typesetting format of latex.

A document forgery detection method comprising using at least one processor for providing at least one histogram of gray level values occurring in at least a portion of at least one channel of an image assumed to represent a document including text, the histogram having been generated by image processing at least a portion of at least one channel of an image assumed to represent a document including text, the image having been sent by a remote end user to an online service over a computer network, evaluating monotony of at least a portion of the at least one histogram; and determining whether the image is authentic or forged based on at least one output of the evaluating.

Systems and methods for mobile enrollment in automated clearing house (ACH) transactions using mobile-captured images of financial documents are provided. Applications running on a mobile device provide for the capture and processing of images of documents needed for enrollment in an ACH transaction, such as a blank check, remittance statement and driver's license. Data from the mobile-captured images that is needed for enrolling in ACH transactions is extracted from the processed images, such as a user's name, address, bank account number and bank routing number. The user can edit the extracted data, select the type of document that is being captured, authorize the creation of an ACH transaction and select an originator of the ACH transaction. The extracted data and originator information is transmitted to a remote server along with the user's authorization so the ACH transaction can be setup between the originator's and receiver's bank accounts.

An image processing apparatus has a color image, the image data being constituted by multiple pixels, each of the multiple pixels having a gradation value, and a controller, which is configured to generate a histogram of index values corresponding to brightness values of the multiple pixels constituting the image data, set an original threshold value based on the histogram which is referred to for binarization, detect a mound-shaped part, in the histogram, satisfying a particular condition, set an adjusting direction in which the original threshold value is to be adjusted, set the index value at a base on a particular direction side of a particular mound-shaped part which is one of mound-shaped parts existing on the adjusting direction side with respect to the original threshold value in the histogram as an adjusted threshold value, and apply a binarizing process to the image data using the adjusted threshold value.

A document type recognition apparatus includes an image region separation unit, a smoothing unit, an edge enhancement unit, a histogram creation unit, and a document type recognition unit. The image region separation unit outputs a signal indicative of each region obtained by separating an input image into a character region and a pattern region. The smoothing unit performs smoothing processing to remove halftone dots of a particular number of lines or greater in the pattern region of the input image. The edge enhancement unit outputs an image subjected to edge enhancement processing depending on an amount of edge on an edge portion of the character region in the input image subjected to the smoothing processing. The histogram creation unit creates a histogram of the image subjected to the edge enhancement processing. The document type recognition unit recognizes a document type of the input image by utilizing the histogram.