Provided are methods and system for recognizing characters such as mathematical expressions or chemical formulas. An example method comprises the steps of receiving and processing an image by a pre-processing module to obtain one or more candidate regions, extracting features of each of the candidate regions by a feature extracting module such as a convolutional neural network (CNN), encoding the features into a distributive representation for each of the candidate regions separately using an encoding module such as a first long short-term memory (LSTM) based neural network, decoding the distributive representation into output representations using a decoding module such as a second LSTM-based recurrent neural network, and combining the output representations into an output expression, which is outputted in a computer-readable format or a markup language.

An apparatus for providing guidance for image capture includes a recognition module that recognizes data objects in data available to an image capture device. At least a portion of the data includes characters. The apparatus includes a context module that determines a context associated with a first image available from the image capture device. The context includes one or more expected data objects, where the first image includes at least a portion of the recognized data objects. The apparatus includes a data module that compares the expected data objects for the determined context, with the recognized data objects from the first image to identify missing expected data objects not found in the first image. The apparatus includes a guidance module that, in response to identified missing data, provides guidance for capturing a second image expected to include the missing expected data objects.

A spectral camera for producing a spectral output is disclosed. The spectral camera has an objective lens for producing an image, an array of mirrors, an array of filters for passing a different passband of the optical spectrum for different ones of the optical channels arranged so as to project multiple of the optical channels onto different parts of the same focal plane, and a sensor array at the focal plane to detect the filtered image copies simultaneously. By using mirrors, there may be less optical degradation and the trade off of cost with optical quality can be better. By projecting the optical channels onto different parts of the same focal plane a single sensor or coplanar multiple sensors can to be used to detect the different optical channels simultaneously which promotes simpler alignment and manufacturing.

Methods and associated systems for generating images with objects of interests positioned at pre-determined locations and in specific orientations (e.g., vertical to the horizon) are disclosed. The method includes generating a preview image by an image component and identifying an object of interest in the preview image. The system then determines a desirable view angle at least based on an orientation of the object of interest. The system generates an original image and a current dip angle based on a measurement performed by a tilt sensor. The system calculates an angle of rotation based on the current dip angle and the desirable view angle. The system then edits or adjusts the original image to form an edited or adjusted image that can be presented to a user.

A method for generating an orientation image, adapted to an electronic device, includes the following steps. A plurality of input pixels including a current input pixel of an input image is received. Edge detection is performed on the current input pixel in a first and second direction to generate a first current gradient and a second current gradient respectively with respect to the first direction and the second direction. A ratio of the first current gradient to a current gradient absolute sum is calculated to obtain a current ratio, where the current gradient absolute sum is a sum of absolute values of the first and second current gradient. A sign of the second current gradient is extracted to obtain a current sign. The current ratio and the current sign are inputted to a lookup table to obtain an image orientation angle corresponding to the current input pixel.

In one example embodiment, an image providing method includes capturing, by a processor, a first partial image while applying a first image processing filter at a first time, capturing, by the processor, a second partial image while applying a second image filter at a second time after the first time, and generating, by the processor, a collage image by merging the first partial image and the second partial image at a third time after the second time.

An image processing method. At least one light ray presented by a vector in a first affine coordinate system with a tip of the probe as an origin is projected from the tip of the probe. The light ray is intercepted from a projection surface satisfying a function in the first affine coordinate. A distance between the tip of the probe and an interception point of the light ray on the projection surface is obtained based on a rotation angle of the probe, a wavelength of the light ray, and a deflection angle of the light ray from the probe. A relationship between the first coordinate and a second affine coordinate system defined with the projection surface as a reference is obtained. Image data are acquired from the light ray reflected from the target surface presented in the first affine coordinate. The image data presented in the first affine coordinate are converted into image data presented in the second affine coordinate, and the image data in the second coordinate system are resampled by interpolating or extrapolating a gray scale.

Visibility of a license plate and color reproducibility of a vehicle body are improved in a monitoring camera.

A vehicle body area detection unit detects a vehicle body area of a vehicle from an image signal. A license plate area detection unit detects a license plate area of the vehicle from the image signal. A vehicle body area image processing unit performs processing of the image signal corresponding to the detected vehicle body area. A license plate area image processing unit performs processing different from the processing of the image signal corresponding to the vehicle body area on the image signal corresponding to the detected license plate area. A synthesis unit synthesizes the processed image signal corresponding to the vehicle body area and the processed image signal corresponding to the license plate area.

A method for detecting a side of an object in an apparatus for detecting a side of an object within an image includes analyzing, by a boundary line detector, an input image to search a detection target region in which a target object is included, detecting, by the boundary line detector, a boundary line between the target object and a ground within the detection target region, calculating, by a side determiner, a gradient for the boundary line, and analyzing, by the side determiner, an image within the detection target region on the basis of the gradient to determine that the entirety of the detection target region is a region of the side of the target object or to determine a separated region corresponding to the side of the target object in the detection target region.

Techniques detecting usage of copyrighted video content using object recognition are provided. In one example, a computer-implemented method comprises determining, by a system operatively coupled to a processor, digest information for a video, wherein the digest information comprises objects appearing in the video and respective times at which the objects appear in the video. The method further comprises comparing, by the system, the digest information with reference digest information for reference videos, wherein the reference digest information identifies reference objects appearing in the reference videos and respective reference times at which the reference objects appear in the reference videos. The method further comprises determining, by the system, whether the video comprises content included in one or more of the reference videos based on a degree of similarity between the digest information and reference digest information associated with one or more of the reference videos.