An image processing device includes: an acquisition unit configured to acquire image data generated by an imaging element where each of narrow band filters constituting a predetermined array pattern is disposed at a position corresponding to any one of a plurality of pixels; a determination unit configured to determine whether or not a light amount of the invisible light range under an environment at the time when the imaging element generates the image data is larger than a threshold value; and a generation unit configured to generate vital information of a subject based on the determination result of the determination unit.

An auto-focus image system that includes a pixel array coupled to a focus signal generator. The pixel array captures an image that has a plurality of edges. The generator generates a focus signal that is a function of a plurality of edge-sharpness measures, each being measured from a different one of the plurality of edges. The edge-sharpness measure is a quantity that has a unit that is a power of a unit of length. It may be a distance in the edge. It may be an area. It may be a central moment. The generator may reduce a relative extent to which an edge contributes to the focus signal on basis of detecting that the edge does not have sufficient reflection symmetry in a sequence of gradients of an image signal across the edge according to a predefined criterion. The edge may be prevented from contributing altogether.

Filtering sensor data is described, for example, where filters conditioned on a local appearance of the signal are predicted by a machine learning system, and used to filter the sensor data. In various examples the sensor data is a stream of noisy video image data and the filtering process denoises the video stream. In various examples the sensor data is a depth image and the filtering process refines the depth image which may then be used for gesture recognition or other purposes. In various examples the sensor data is one dimensional measurement data from an electric motor and the filtering process denoises the measurements. In examples the machine learning system comprises a random decision forest where trees of the forest store filters at their leaves. In examples, the random decision forest is trained using a training objective with a data dependent regularization term.

Enhanced defect detection of a sample includes acquiring two or more inspection images from a sample from two or more locations of the sample for a first optical mode. The defect detection also generates an aggregated defect profile based on the two or more inspection images from the two or more locations for the first optical mode for a selected defect type and calculating one or more noise correlation characteristics of the two or more inspection images acquired from the two or more locations for the first optical mode. Defect detection further includes the generation of a matched filter for the first optical mode based on the generated aggregated defect profile and the calculated one or more noise correlation characteristics.

In one embodiment of the present invention, a native resolution analyzer generates a log-magnitude spectrum that elucidates sampling operations that have been performed on a scene. In operation, the native resolution analyzer performs a transform operation of a color component associated with a frame included in the scene to generate a frame spectrum. The native resolution analyzer then normalizes the magnitudes associated with the frame spectrum and logarithmically scales the normalized magnitudes to create a log-magnitude frame spectrum. This two dimensional log-magnitude frame spectrum serves as a frequency signature for the frame. More specifically, patterns in the log-magnitude spectrum reflect re-sampling operations, such as a down-sampling and subsequent up-sampling, that may have been performed on the frame. By analyzing the log-magnitude spectrum, discrepancies between the display resolution of the scene and the lowest resolution with which the scene has been processed may be detected in an automated fashion.

A three-dimensional object detection has an image capturing device, an image conversion unit, a three-dimensional object detection unit, a three-dimensional object assessment unit, first and second foreign matter detection units and a controller. The image capturing device captures images rearward of a vehicle. The three-dimensional object detection unit detects three-dimensional objects based on image information. The three-dimensional object assessment unit assesses whether or not a detected three-dimensional object is another vehicle. The foreign matter detection units detect whether or not foreign matter has adhered to a lens based on the change over time in luminance values for each predetermined pixel of the image capturing element and the change over time in the difference between an evaluation value and a reference value. The controller outputs control commands to the other means to suppress the assessment of foreign matter as another vehicle when foreign matter has been detected.

A control device includes a flatness calculator and control-value logic. The flatness calculator calculates flatness of an image corresponding to an image signal. The control-value logic determines a control value based on the flatness. The control value controls a size of an operation region on which a first-order infinite impulse response (IIR) filter is to process the image signal to reduce noise.

A multi-scale detail representation of an image is computed as a weighted sum of translation difference images. A denoising operator is applied to the translation difference images so that translation differences are modified as a function of an estimated local signal-to-noise ratio and at least one denoised center difference image at a specific scale is computed by combining denoised translation difference images at scale s or a finer scale. A denoised image is computed by applying a reconstruction algorithm to the denoised center difference images.

Because neighboring frames may affect how a current frame is perceived, we examine different neighborhoods of the current frame and select a neighborhood that impacts the perceived temporal distortion (i.e., when frames are viewed continuously) of the current frame most significantly. Based on spatial distortion (i.e., when a frame is viewed independently of other frames in a video sequence) of frames in the selected neighborhood, we can estimate initial temporal distortion. To refine the initial temporal distortion, we also consider the distribution of distortion in the selected neighborhood, for example, the distance between the current frame and a closest frame with large distortion, or whether distortion occurs in consecutive frames.

A quality of an image obtained by image processing is improved with a shorter amount of time spent in the image processing. An image processing apparatus includes, an acquisition unit configured to acquire an image, a determination unit configured to determine a quality of the acquired image, a switching unit configured to switch between settings of acquiring and not acquiring a reference image, from a communication apparatus connected to the image processing apparatus, used for improving the quality of the acquired image, based on the determined quality, and a processing unit configured to perform image processing by using the reference image acquired from the communication apparatus, in accordance with the switching by the switching unit.