An image processing apparatus includes a first composition unit which generates a first HDR image by applying a first gamma to each of a plurality of images different in exposure amount and composing the plurality of images after the application of the first gamma, a determination unit which determines whether one preset image among the plurality of images includes a light region satisfying a preset condition, a generation unit which generates, based on the light region, map data for discriminating the light region, a dark region, and an intermediate region, and a second composition unit which generates a second HDR image by applying a second gamma to one of the plurality of images, and composing, with reference to the map data, an image obtained by applying the second gamma and the first HDR image data.

A system for simplified generation of systems for analysis of satellite images to geolocate one or more objects of interest. A plurality of training images labeled for a study object or objects with irrelevant features loaded into a preexisting feature identification subsystem causes automated generation of models for the study object. This model is used to parameterize pre-engineered machine learning elements that are running a preprogrammed machine learning protocol. Training images with the study are used to train object recognition filters. This filter is used to identify the study object in unanalyzed images. The system reports results in a requestor's preferred format.

A web content enrichment system can match an image to text of web content. When the text of web content includes a snippet, the image matched to the text enriches the snippet to enhance results of a search engine. When the text of web content includes text contained in a webpage, the image matched to this text enriches the webpage to enhance user perception and understanding of the webpage. The process of matching images to text involves extracting features of a plurality of images and features of a plurality of text documents, calculating scores of the images based on the extracted features, and selecting one image per text document based on the scores using a machine-learning algorithm. The result of the matching can be provided to a web content module for storing, incorporating into the result lists of the search engine, or delivery to a user.

Depths of one or more objects in a scene may be measured with enhanced accuracy through the use of a light-field camera and a depth sensor. The light-field camera may capture a light-field image of the scene. The depth sensor may capture depth sensor data of the scene. Light-field depth data may be extracted from the light-field image and used, in combination with the sensor depth data, to generate a depth map indicative of distance between the light-field camera and one or more objects in the scene. The depth sensor may be an active depth sensor that transmits electromagnetic energy toward the scene; the electromagnetic energy may be reflected off of the scene and detected by the active depth sensor. The active depth sensor may have a 360° field of view; accordingly, one or more mirrors may be used to direct the electromagnetic energy between the active depth sensor and the scene.

An image recognition method is disclosed. The method includes acquiring an image detecting image information and a position of a polygon object included in the image; projecting the image information of the polygon object onto the recognition area based on the position of the polygon object and a position of a recognition area to obtain a projection image; and recognizing the projection image using an image recognition technology to obtain information in the polygon object. Projecting image information of a polygon object onto a recognition area and performing recognition thereon are equivalent to correcting a shape and a position of the polygon object in the recognition area, such that an image after the correction can be recognized. As such, a failure in recognition due to a failure of a position, a shape and the like of a polygon object in a recognition area in fulfilling the recognition requirements is solved.

Image data relating to real-world objects or persons is collected from a scene while collecting audio data relating to the real-world objects or persons from the same scene. The audio data is used to derive sound objects corresponding to the real-world objects or persons. The image data is used to derive video objects corresponding to the real-world objects or persons. Based on the sound objects and the video objects, candidate salient objects are generated. A salient object is selected from among the candidate salient objects. Perceptual enhancement operations are performed on the selected salient object.

Among other things, one or more systems and/or techniques for segmenting a representation of a sheet object from an image are provided herein. To identify elements of an image (e.g., pixels and/or voxels) representative of sheet objects, a constant false alarm rate (CFAR) score and a topological score are computed for respective elements being analyzed. The CFAR score indicates a relationship between an element and a neighborhood of elements when viewed as a collective unit. The topological score indicates a relationship between the element and a neighborhood of elements when viewed neighbor-by-neighbor. When the CFAR score is within a specified range of CFAR scores and the topological score is within a specified range of topological scores, the element is labeled as being associated with a sheet object. A connected component labeling (CCL) approach may be used to group elements labeled as being associated with a sheet object.

An image generating device has a camerawork information extracting unit that extracts imaging information of a captured image, a composition information calculating unit that decides a cut-out frame for cutting out a new image from the imaging information and the captured image based on a constraint condition, and a composition information shaping unit that outputs, as attribute information of the captured image, the imaging information and the cut-out frame. This configuration makes it possible to obtain from a captured image a newly generated image which is a favorable image that is easy to see for a viewer, without requiring skilled imaging techniques during capturing, and to efficiently record and manage the captured image and the newly generated image.

An image recognition system and method for a vehicle, including at least two camera units, each being configured to record an image of a road in the vicinity of the vehicle and to provide image data representing the respective image of the road, a first image processor configured to combine the image data provided by the at least two camera units into a first top-view image. The first top-view image is aligned to a road image plane, a first feature extractor configured to extract lines from the first top-view image, a second feature extractor configured to extract an optical flow from the first top-view image and a second top-view image, generated before the first top-view image by the first image processor, and a curb detector configured to detect curbs in the road based on the extracted lines and the extracted optical flow and provide curb data representing the detected curbs.

A fast object detection method based on deformable part model (DPM) is provided. The method includes importing a trained classifier for object detection, receiving an image frame from a plurality of frames in a video captured by a camera, and identifying regions possibly containing at least one object via objectness measure based on Binarized Normed Gradients (BING). The method also includes calculating Histogram of Oriented Gradients (HOG) feature pyramid of the image frame, performing DPM detection for the identified regions possibly containing the at least one object, and labeling the at least one detected object using at least one rectangle box via non-maximum suppression (NMS). Further, the method includes processing a next frame from the plurality of frames in the captured video until the video ends and outputting object detection results.