Abstract: The disclosure provides a document processing apparatus, method and a scanner. The document processing apparatus includes: a text line extraction unit extracting a text line from an input document; a language classification unit determining whether an OCR process is necessary for a language of the input document; an OCR unit determining, by performing the OCR process, an OCR confidence in the case that it is determined that the OCR process is necessary; an graphic feature recognition unit determining an graphic feature recognition confidence; and a determination unit determining a combination confidence based on at least one of the determined graphic feature recognition confidences and the determined OCR confidences, and determining an orientation of the input document based on the combination confidences. This technical solution can determine better an orientation of the document, and is especially applicable when the quality of the image of the document is deteriorated.

Abstract: Systems and methods are disclosed for autonomous driving with only a single camera by moving object localization in 3D with a real-time framework that harnesses object detection and monocular structure from motion (SFM) through the ground plane estimation; tracking feature points on moving cars a real-time framework to and use the feature points for 3D orientation estimation; and correcting scale drift with ground plane estimation that combines cues from sparse features and dense stereo visual data.

Abstract: The present invention provides a large format fingerprint capture apparatus, system and method that is low power, compact, and lightweight and has a platen area greater than 3.0 square inches. The present system is typically powered, controlled, and exchanges data over a single data/control/power connection to a host PC, e.g., a desk top computer, PDA, or laptop computer although the system can also be used in a wireless fashion with a power subsystem so no physical connections are required. The system typically includes a light source, a prism, a camera (including the lens), and a case. Optional elements comprise holographic elements such as gratings and holographic optical elements (HOEs), a battery subsystem, magnetic stripe reader, barcode reader, platen heater, platen blower, and mirrors to divert the image beam.

Abstract: Provided is a Fingerprint sensor using Acoustic Impediography. The sensor includes an Application Specific Integrated Circuit (ASIC or IC) and an array of mechanical resonator used as sensing elements. The array of sensing elements contains multiple sensing elements arranged in rows and columns.

Abstract: Aspects of the present invention comprise generating and using Multi-Order Contextual co-Occurrence (MOCO) descriptors to implicitly model the high level context using detection responses from a baseline object detector. In embodiments, a 1st-order context feature is computed as a set of randomized binary comparisons on a response map of the baseline object detector. The statistics of the 1st-order binary context features are further calculated to construct a higher-order co-occurrence descriptor, which, in embodiments, may be combined with other features such as the 0th-order context features and/or the 1st-order features to form the MOCO. In embodiments, combining the MOCO feature with the original image feature, the baseline object detector may be evolved to a stronger context aware detector.

Abstract: A method for replacing image data in a destination region that is divided into sub-pieces along one or more cutting paths, which start and end at two different points on the border, and finding replacement data for the sub-pieces. The cutting paths may be determined as a function of the type of image structured at the start and the end points. The cutting paths may also be determined as a function of the area of the sub-pieces and the lengths of the cutting paths. Optionally, the destination region may be determined by a spot detection algorithm. Further optionally, the spot detection algorithm may comprise calculation of a high pass filter, or detection of areas of luminosity and border-to-volume ratios. A method for moving an image element within an image is also provided.

Abstract: Systems and methods are disclosed for object detection by receiving an image and extracting features therefrom; applying a learning process to determine sub-regions and select predetermined pooling regions; and performing selective max-pooling to choose one or more feature regions without noises.

Abstract: A method for automatically tracking multiple objects from a sequence of video images that may extract raw data about participating elements in a sporting, or other event, in a way that does not interfere with the actual participating elements in the event. The raw data may include the position and velocity of the players, the referees, and the puck, as well as the team affiliation of the players. These data may be collected in real time and may include accounting for players moving fast and unpredictably, colliding with and occluding each other, and getting in and out of the playing field. The video sequence, captured by a suitable sensor, may be processed by a suitably programmed general purpose computing device.

Abstract: A method of determining relief markings on a tire's sidewall surface includes assigning, to each pixel of a three-dimensional image of the surface, a grey-level value proportional to an elevation point corresponding to the pixel, to obtain a starting image. Using linear structuring elements of successively increasing sizes and oriented circumferentially, a series of successive morphological openings is performed iteratively on the starting surface. An image value obtained after a morphological opening using a structuring element is subtracted from an image value obtained after a morphological opening using a structuring element of an immediately lower size, to obtain a succession of images flattened by differencing. A thresholding operation is performed on the images flattened by differencing, to obtain binary images. A set-theoretic union of values of each of the binary images is performed, to obtain a final binary image in which markings appear in relief.

Abstract: Systems and methods are disclosed for object detection by receiving an image; segmenting the image; extracting features from the image; and performing a dimension-wise spatial layout selection to pick up dimensions inside a discriminative spatial region for classification.

Abstract: Systems and methods for object detection by receiving an image; segmenting the image and identifying candidate bounding boxes which may contain an object; for each candidate bounding box, dividing the box into overlapped small patches, and extracting dense features from the patches; during a training phase, applying a learning process to learn one or more discriminative classification models to classify negative boxes and positive boxes; and during an operational phase, for a new box generated from the image, applying the learned classification model to classify whether the box contains an object.

Abstract: A computer-implemented method for creating an ordered set of boundary data by transforming data from remotely sensed imagery of shorelines is provided. A feature data set and an edge data set are transformed into a set of 3-point boundary segments having a specific head and tail point and the segments are ordered from tail to head in a clockwise or counterclockwise manner relative to the water. Once the 3-point segments are created they are easily linked together into larger segments. These large multi-point segments in turn are linked together to create a closed loop in a predetermined direction, for example, but not limited to, the shorelines for rivers or coastal areas.

Abstract: Foreground object image features are extracted from input video via application of a background subtraction mask, and optical flow image features from a region of the input video image data defined by the extracted foreground object image features. If estimated movement features indicate that the underlying object is in motion, a dominant moving direction of the underlying object is determined. If the dominant moving direction is parallel to an orientation of the second, crossed thoroughfare, an event alarm indicating that a static object is blocking travel on the crossing second thoroughfare is not generated. If the estimated movement features indicate that the underlying object is static, or that its determined dominant moving direction is not parallel to the second thoroughfare, an appearance of the foreground object region is determined and a static-ness timer run while the foreground object region comprises the extracted foreground object image features.

Abstract: A method of determining relief elements on a tire surface includes capturing a three-dimensional image of the surface and assigning to each pixel (i, j) of the image a grey-level value proportional to a topographic elevation of a point corresponding to the pixel, so as to obtain a starting image (ƒ). The method also includes transforming the starting image (ƒ) by a first top-hat operation, with aid of a first structuring element, so as to obtain an image (ƒ1) flattened by differencing, (ƒ1=ƒ—?3,150(ƒ)). The method further includes performing a thresholding operation on the image flattened by differencing, so as to obtain a segmented image, (T<1,30>(ƒ1)=seg1) in which relief elements appear white on a black background.

Abstract: Automatic object retrieval from input video is based on learned, complementary detectors created for each of a plurality of different motionlet clusters. The motionlet clusters are partitioned from a dataset of training vehicle images as a function of determining that vehicles within each of the scenes of the images in each cluster share similar two-dimensional motion direction attributes within their scenes. To train the complementary detectors, a first detector is trained on motion blobs of vehicle objects detected and collected within each of the training dataset vehicle images within the motionlet cluster via a background modeling process; a second detector is trained on each of the training dataset vehicle images within the motionlet cluster that have motion blobs of the vehicle objects but are misclassified by the first detector; and the training repeats until all of the training dataset vehicle images have been eliminated as false positives or correctly classified.

Abstract: Methods, apparatus, and articles of manufacture to measure geographical features using an image of a geographical location are disclosed. An example method includes dividing, with a processor, an image of a geographic area of interest into a plurality of geographical zones, the geographical zones being representative of different geographical areas having approximately equal physical areas, measuring, with the processor, a geographical feature represented in the image for corresponding ones of the plurality of geographical zones, storing descriptions for the geographical zones in a computer memory, and storing values representative of the geographical feature of the geographical zones.

Abstract: Disclosed are a system for making 3D contents provided with visual fatigue minimization and a method of the same. More particularly, an exemplary embodiment of the present invention provides a system for making 3D contents including: a human factor information unit generating guide information for making 3D contents by considering factors causing visual fatigue of the 3D contents; and a 3D contents making unit applying guide information generated by the human factor information unit to 3D contents data inputted for making the 3D contents to make the 3D contents, and a method of making 3D contents.

Abstract: A method of conducting pattern matching is provided that includes establishing probe categories. Each probe category corresponds to pattern characteristics of one of a plurality of subpopulations. Moreover, the method includes coordinating combinations of the subpopulations and probe categories with pattern matching systems such that each combination corresponds to at least one of a plurality of the pattern matching systems, obtaining pattern data for an object, configuring the obtained object pattern data as a probe, and determining the probe category of the probe. Furthermore, the method includes conducting a matching transaction between the probe and each of the subpopulations using the at least one matching system corresponding to each combination of subpopulation and the determined probe category, and determining at least one candidate match when the probe matches at least one enrollment data record in the at least one matching system of any of the subpopulations.

Abstract: A method of detecting the smoke of a forest fire using the spatiotemporal Bag-of-Features (BoF) of the smoke and a random forest is provided. In the method, whenever each frame of a video sequence is input, a difference between the input frame and a previous frame is detected, and the input frame is set as a key frame if the difference exceeds a predetermined first threshold value. One or more moving blocks are detected in the set key frame. One or more candidate smoke blocks are extracted from the moving blocks using a smoke color model. BoF representations are generated from the detected candidate smoke blocks. Whether smoke of the candidate smoke blocks is actual smoke is determined by performing random forest learning on the generated BoF representation.

Abstract: A method and an apparatus of image depth estimation are provided. The method includes the following steps. First, a hue value of each pixel in an image is calculated by comparing all color components of each pixel in the image. The hue value of each pixel in the image is associated with a corresponding value, wherein the corresponding value is a first numerical value or a second numerical value. Then, according to the corresponding value of each pixel in the image, a depth value of each pixel in the image is calculated, in which the depth value is used to convert the image into a three-dimensional (3D) image to be displayed on a 3D display apparatus.