A system and method are described for enhancing readability of document images by operating on each document individually. Monochromatic light sources operating at different wavelengths of light can be used to obtain greyscale images. The greyscale images can then be used in any desired image enhancement algorithm. In one example algorithm, an automated method removes image background noise and improves sharpness of the scripts and characters using edge detection and local color contrast computation.

Methods and systems for image binarization using mean restrain are disclosed. A method includes: obtaining, by a computing device, a grayscale image; generating, by the computing device, a histogram from the grayscale image; determining, by the computing device, a foreground mean value and a background mean value for each pixel value in the histogram; determining, by the computing device, a binarization threshold using the foreground mean values and the background mean values; and generating, by the computing device, a binarized image using the grayscale image and the binarization threshold.

An alignment method for manufacturing a mask integration framework is disclosed. The alignment method includes establishing an absolute coordinate system by taking a center of a metal framework as an origin of coordinates, the center of the metal framework coinciding with a center of an array substrate serving as a reference, controlling the array substrate to move, such that an offset of coordinates of a pixel point under the absolute coordinate system with respect to a predetermined theoretical value is smaller than or equal to a predetermined error value, and transmitting the coordinates of the pixel point under the absolute coordinate system, after the array substrate moves, to a tension device. An alignment system for manufacturing mask integration framework is also disclosed.

A sensing method for a nanostructure pixel sensor includes shining light from an incident angle onto a nanostructure pixel sensor; capturing an image of nanostructure pixels of the nanostructure pixel sensor; obtaining brightness values of the nanostructure pixels from the image; and determining a presence and/or amount of an analyte based on an image pattern. The nanostructure pixel sensor includes a plurality of the nanostructure pixels. Each of the nanostructure pixels includes periodic nanostructures. At least two nanostructure pixels have different periodic spacings, and the analyte is applied on the nanostructure pixel sensor before capturing the image.

Described herein are systems, methods, and computer-readable media for generating and training a high precision low bit convolutional neural network (CNN). A filter of each convolutional layer of the CNN is approximated using one or more binary filters and a real-valued activation function is approximated using a linear combination of binary activations. More specifically, a non-11 filter (e.g., a kk filter, where k>1) is approximated using a scaled binary filter and a 11 filter is approximated using a linear combination of binary filters. Thus, a different strategy is employed for approximating different weights (e.g., 11 filter vs. a non-11 filter). In this manner, convolutions performed in convolutional layer(s) of the high precision low bit CNN become binary convolutions that yield a lower computational cost while still maintaining a high performance (e.g., a high accuracy).

Various embodiments disclosed herein are directed to methods of capturing Vehicle Identification Numbers (VIN) from images captured by a mobile device. Capturing VIN data can be useful in several applications, for example, insurance data capture applications. There are at least two types of images supported by this technology: (1) images of documents and (2) images of non-documents.

An online system is configured to provide content items to users. The content item includes an image, and is displayed with an interface element colored using an accent color to create a unified look and feel with the displayed image. The accent color is dynamically selected based upon the image, extracted color features of the image, and embeddings associated with the image indicating at least one object depicted in the image. A machine-trained classification model selects the color to be used in displaying the interface element from a quantized set of colors of the image, based upon the extracted color features and the embeddings associated with the image. As such, suitable accent colors can be selected automatically for large numbers of content items, in a flexible manner that can account for the context of the images and the context in which the content item is to be displayed.

A sensing method for a nanostructure pixel sensor includes shining light from an incident angle onto a nanostructure pixel sensor; capturing an image of nanostructure pixels of the nanostructure pixel sensor; obtaining brightness values of the nanostructure pixels from the image; and determining a presence and/or amount of an analyte based on an image pattern. The nanostructure pixel sensor includes a plurality of the nanostructure pixels. Each of the nanostructure pixels includes periodic nanostructures. At least two nanostructure pixels have different periodic spacings, and the analyte is applied on the nanostructure pixel sensor before capturing the image.

Systems, devices, media, and methods are presented for receiving a set of images in a video stream, converting one or more images of the set of images to a set of single channel images, generating a set of approximation images from the set of single channel images, and generating a set of binarized images by thresholding the set of approximation images.

The present disclosure provides a QR code graphic manufacturing device, which comprises a gray-scaled adjusting module, a data-hiding module, and an error-diffusion module. The gray-scaled adjusting module is used to adjust a gray-scaled value of an original image to generate a modified gray-scaled image. The data-hiding module and the error-diffusion module are respectively used to process the modified gray-scaled image and a QR code, and to generate a QR code graphic with a better hidden effect.