A method for stray light compensation is disclosed. The method comprising: acquiring a first image with a first imaging device covering a first field-of-view; acquiring a second image with a second imaging device covering a second field-of-view, wherein the second field-of-view is larger than the first field-of-view and wherein the first field-of-view is included in the second field-of-view; estimating stray light components in pixels of the first image from pixel data of pixels in the second image; and compensating for stray light in the first image by subtracting the estimated stray light components in pixels of the first image. Also, a system for stray light compensation is disclosed.

Methods and apparatus associated with predicting colorectal cancer tumor invasiveness are described. One example apparatus includes a set of circuits, and a data store that stores radiological images of tissue demonstrating colorectal cancer. The set of circuits includes a circumferential resection margin (CRM) prediction circuit that generates a CRM probability score for a diagnostic radiological image, an image acquisition circuit that acquires a diagnostic radiological image of a region of tissue demonstrating colorectal cancer pathology and that provides the diagnostic radiological image to the CRM prediction circuit, and a training circuit that trains the CRM prediction circuit to quantify chemoradiation response in the region of tissue represented in the diagnostic radiological image. The training circuit trains the CRM prediction circuit using a set of composite images.

Embodiments of a computer-implemented method for training a convolutional neural network (CNN) that is pre-trained using a set of color images are disclosed. The method comprises receiving a training dataset including multiple multidimensional images, each multidimensional image including a color image and a depth image; performing a fine-tuning of the pre-trained CNN using the depth image for each of the plurality of multidimensional images; obtaining a depth CNN based on the pre-trained CNN, wherein the depth CNN is associated with a first set of parameters; replicating the depth CNN to obtain a duplicate depth CNN being initialized with the first set of parameters; and obtaining a depth-enhanced color CNN based on the duplicate depth CNN being fine-tuned using the color image for each of the plurality of multidimensional images, wherein the depth-enhanced color CNN is associated with a second set of parameters.

Techniques disclosed herein utilize a vision sensor that integrates a special-purpose camera with dedicated computer vision (CV) computation hardware and a dedicated low-power microprocessor for the purposes of detecting, tracking, recognizing, and/or analyzing subjects, objects, and scenes in the view of the camera. The vision sensor processes the information retrieved from the camera using the included low-power microprocessor and sends events (or indications that one or more reference occurrences have occurred, and, possibly, associated data) for the main processor only when needed or as defined and configured by the application. This allows the general-purpose microprocessor (which is typically relatively high-speed and high-power to support a variety of applications) to stay in a low-power (e.g., sleep mode) most of the time as conventional, while becoming active only when events are received from the vision sensor.

A method for detecting deviation from a preferred cervical spine posture when using a mobile device is disclosed. The mobile device uses a front-facing camera to capture images of the user and apply a nostril tracking algorithm to the images. The nostril tracking algorithm is used in real-time to measure displacement of the user's nostrils and correlate the nostril displacement to a cervical spine flexion angle. The user's cervical spine flexion angle is communicated using an alarm device, such as a row of lights, which allows the user to monitor and correct their posture and avoid potential injury.

A computer-implemented technique for feature extraction includes obtaining an electronic image of an object and performing an edge detection algorithm on the electronic image. The technique further includes performing an edge pooling algorithm and sampling the electronic image edge patches, color patches and texture patches. A set of patches is selected from the edge patches, color patches and texture patches by selecting an (i.sup.th+1) patch to be within the set of patches based on a Euclidean distance from an i.sup.th patch of the set of patches for each of the set of edge patches, the set of color patches and the set of texture patches. A part selection algorithm and a part pooling algorithm is performed to obtain parts that are registered to the object.

A method is disclosed including: receiving raw image data corresponding to a series of raw images; processing the raw image data with an encoder to generate encoded data, where the encoder is characterized by an input/output transformation that substantially mimics the input/output transformation of one or more retinal cells of a vertebrate retina; and applying a first machine vision algorithm to data generated based at least in part on the encoded data.

A biometrics authentication device is configured to include a non-directional feature generation process unit configured to generate a non-directional feature on the basis of a directional features; a directional feature generation process unit configured to select, from among the directional features, a reference directional feature corresponding to a reference direction; a non-directional feature matching process unit configured to obtain a first degree of similarity between the non-directional feature and a registered non-directional feature; a directional feature matching process unit configured to obtain a second degree of similarity between the reference directional feature and a registered reference directional feature; and a determination unit configured to make a weight of the second degree of similarity smaller than a weight of the first degree of similarity and to determine whether or not a subject is a person to be authenticated, by using the first degree of similarity and the second degree of similarity.

A method and apparatus for assisting a vehicle to stop at crosswalks and intersections by using a surround view monitor (SVM). In particular, the method includes: obtaining a front top-view image of the vehicle by using the SVM, performing a stop line detection in the front top-view image, and performing a crosswalk detection in the front top-view image in response to no detection of a stop line. The method further includes: generating a virtual stop line in response to a detection of a crosswalk, and controlling the vehicle to stop based on the detected stop line or the generated virtual stop line.

Disclosed is an image analysis method including inputting analysis data into a classifier having a neural network structure, the analysis data being generated from an image of an analysis target cell and including information regarding the analysis target cell; and classifying, by use of the classifier, the analysis target cell into at least one of categories of a blood cell. The categories include morphological features of white blood cells. The morphological features of white blood cells include at least morphological nucleus abnormality, presence of vacuole, granule morphological abnormality, granule distribution abnormality, presence of abnormal granule, cell size abnormality, presence of inclusion body, or bare nucleus.