3280 results about "Object detection" patented technology

A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of a first object detected in a first group of the images. The first object has a first signature representing features of the first object. The match classifier matches object instances by analyzing data derived from the first signature of the first object and a second signature of a second object detected in a second image. The second signature represents features of the second object derived from the second image. The match classifier determine whether the second signature matches the first signature. A training process automatically configures the match classifier using a set of possible object features.

An object recognition apparatus that can recognize an object in an image acquired by an on-vehicle camera mounted on a vehicle with high efficiency and accuracy. A recognition processing is carried out in a cut off range, which is cut off from the image, and in which an object to be recognized is supposed to be present. The cut off range is obtained using support information supplied from a position detection device and/or an anti-lock brake device of the vehicle. Since the object in the image changes its position as the vehicle moves, the displacement must be estimated during the object recognition. The estimation requires camera attitude parameters of the on-vehicle camera with respect to the road, and the estimation is performed using support information supplied from the position detection device and/of the anti-lock brake device. This will improve the efficiency and accuracy of obtaining the camera attitude parameters.

A vehicle awareness system for monitoring remote vehicles relative to a host vehicle. The vehicle awareness system includes at least one object sensing device and a vehicle-to-vehicle communication device. A data collection module is provided for obtaining a sensor object data map and vehicle-to-vehicle object data map. A fusion module merges the sensor object data map and vehicle-to-vehicle object data map for generating a cumulative object data map. A tracking module estimates the relative position of the remote vehicles to the host vehicle.

A method and system for anatomical object detection using marginal space deep neural networks is disclosed. The pose parameter space for an anatomical object is divided into a series of marginal search spaces with increasing dimensionality. A respective sparse deep neural network is trained for each of the marginal search spaces, resulting in a series of trained sparse deep neural networks. Each of the trained sparse deep neural networks is trained by injecting sparsity into a deep neural network by removing filter weights of the deep neural network.

A wireless energy transfer system includes a foreign object debris detection system. The system includes at least one wireless energy transfer source configured to generate an oscillating magnetic field. The foreign object debris may be detected by at least one field gradiometer positioned in the oscillating magnetic field. The voltage of the at least one field gradiometer may be measured using readout circuitry and a feedback loop based on the readings from the gradiometers may be used to control the parameters of the wireless energy source.

An imaging system for a vehicle includes an imaging array sensor and a control. The image array sensor comprises a plurality of photo-sensing pixels and is positioned at the vehicle with a field of view exteriorly of the vehicle. The imaging array sensor is operable to capture an image of a scene occurring exteriorly of the vehicle. The captured image comprises an image data set representative of the exterior scene. The control algorithmically processes the image data set to a reduced image data set of the image data set. The control processes the reduced image data set to extract information from the reduced image data set. The control selects the reduced image data set based on a steering angle of the vehicle.

Aspects of the present disclosure relate generally to modeling a vehicle's view of its environment. This view need not include what objects or features the vehicle is actually seeing, but rather those areas that the vehicle is able to observe using its sensors if the sensors were completely un-occluded. For example, for each of a plurality of sensors of the object detection component, a computer may an individual 3D model of that sensor's field of view. Weather information is received and used to adjust one or more of the models. After this adjusting, the models may be aggregated into a comprehensive 3D model. The comprehensive model may be combined with detailed map information indicating the probability of detecting objects at different locations. A model of the vehicle's environment may be computed based on the combined comprehensive 3D model and detailed map information and may be used to maneuver the vehicle.

A video surveillance system uses rule-based reasoning and multiple-hypothesis scoring to detect predefined behaviors based on movement through zone patterns. Trajectory hypothesis spawning allows for trajectory splitting and/or merging and includes local pruning to managed hypothesis growth. Hypotheses are scored based on a number of criteria, illustratively including at least one non-spatial parameter. Connection probabilities computed during the hypothesis spawning process are based on a number of criteria, illustratively including object size. Object detection and probability scoring is illustratively based on object class.

A method and system for anatomical object detection using marginal space deep neural networks is disclosed. The pose parameter space for an anatomical object is divided into a series of marginal search spaces with increasing dimensionality. A respective deep neural network is trained for each of the marginal search spaces, resulting in a series of trained deep neural networks. Each of the trained deep neural networks can evaluate hypotheses in a current parameter space using discriminative classification or a regression function. An anatomical object is detected in a medical image by sequentially applying the series of trained deep neural networks to the medical image.

A method for determining the orientation of a digital image, includes the steps of: employing a semantic object detection method to detect the presence and orientation of a semantic object; employing a scene layout detection method to detect the orientation of a scene layout; and employing an arbitration method to produce an estimate of the image orientation from the orientation of the detected semantic object and the detected orientation of the scene layout.

The present invention detects a moving object by generating the distance information of the moving object, detecting the object motion, determining the object distance, detecting the object image area and the object contour from the video image that includes the object image and contour, and provides a moving object detection apparatus to carry out such detection as well as detecting a contour of the specific moving object by detecting the center of the moving object in high precision.

A computer-implemented method executed by at least one processor for training fast models for real-time object detection with knowledge transfer is presented. The method includes employing a Faster Region-based Convolutional Neural Network (R-CNN) as an objection detection framework for performing the real-time object detection, inputting a plurality of images into the Faster R-CNN, and training the Faster R-CNN by learning a student model from a teacher model by employing a weighted cross-entropy loss layer for classification accounting for an imbalance between background classes and object classes, employing a boundary loss layer to enable transfer of knowledge of bounding box regression from the teacher model to the student model, and employing a confidence-weighted binary activation loss layer to train intermediate layers of the student model to achieve similar distribution of neurons as achieved by the teacher model.

An interruption free navigator includes an inertial measurement unit, a north finder, a velocity producer, a positioning assistant, a navigation processor, an altitude measurement, an object detection system, a wireless communication device, and a display device and map database. Output signals of the inertial measurement unit, the velocity producer, the positioning assistant, the altitude measurement, the object detection system, and the north finder are processed to obtain highly accurate position measurements of the person. The user's position information can be exchanged with other users through the wireless communication device, and the location and surrounding information can be displayed on the display device by accessing a map database with the person position information.

Object recognition apparatus and methods useful for extracting information from an input signal. In one embodiment, the input signal is representative of an element of an image, and the extracted information is encoded into patterns of pulses. The patterns of pulses are directed via transmission channels to a plurality of detector nodes configured to generate an output pulse upon detecting an object of interest. Upon detecting a particular object, a given detector node elevates its sensitivity to that particular object when processing subsequent inputs. In one implementation, one or more of the detector nodes are also configured to prevent adjacent detector nodes from generating detection signals in response to the same object representation. The object recognition apparatus modulates properties of the transmission channels by promoting contributions from channels carrying information used in object recognition.

A vehicle has a plurality of control apparatuses, a user input, a geographic position component, an object detection apparatus, memory, and a display. A processor is also included and is programmed to receive the destination information, identify a route, and determine the current geographic location of the vehicle. The processor is also programmed to identify an object and object type based on object information received from the object detection apparatus and to determine at least one warning characteristic of the identified object based on at least one of: the object type, a detected proximity of the detected object to the vehicle, the location of the detected object relative to predetermined peripheral areas of the vehicle, the current geographic location of the vehicle, and the route. The processor is also configured to select and display on the display an object warning image based on the at least one warning characteristic.

PCT No. PCT/US95/02155 Sec. 371 Date Dec. 22, 1997 Sec. 102(e) Date Dec. 22, 1997 PCT Filed Feb. 21, 1995 PCT Pub. No. WO95/22804 PCT Pub. Date Aug. 24, 1995A method and ultra-compact system has been developed for illuminating and detecting the surface topography of an object such as the finger (4) of an individual. The system (8) is capable of producing high-contrast images which can be electronically transmitted in real-time, or stored using electronic or photographic recording devices. Light traveling within a light transmitting substrate (2) is redirected by a slanted-fringed light diffractive grating preferably embodied within a volume hologram (3). The volume hologram (3), either of the reflection or transmission type, is attached to the light transmitting substrate (2). and functions to diffract light striking thereupon and illuminate an object having topographical surface structure. After being spatially and intensity modulated in accordance with topographical details of the illuminated object, the insulated light passes back through the light transmitting substrate (2) and the volume hologram (3), onto an image detection array. for subsequent analysis. Each of the disclosed embodiments has a compact geometry suitable for use in diverse object identification applications.

Object recognition apparatus and methods useful for extracting information from sensory input. In one embodiment, the input signal is representative of an element of an image, and the extracted information is encoded in a pulsed output signal. The information is encoded in one variant as a pattern of pulse latencies relative to an occurrence of a temporal event; e.g., the appearance of a new visual frame or movement of the image. The pattern of pulses advantageously is substantially insensitive to such image parameters as size, position, and orientation, so the image identity can be readily decoded. The size, position, and rotation affect the timing of occurrence of the pattern relative to the event; hence, changing the image size or position will not change the pattern of relative pulse latencies but will shift it in time, e.g., will advance or delay its occurrence.

In a robot control apparatus mounted on a mobile robot, movement of a human existing in front of the robot is detected, and the robot is moved in association with the movement of the human to thereby obtain path teaching data. When the robot moves autonomously according to the path teaching data, a robot movable area with respect to the path teaching data is calculated from positions of the ceiling and walls of the robot moving space or positions of obstacles detected by a surrounding object detection unit, whereby a moving path for autonomous movement is generated. The robot is controlled to move autonomously by a drive of a drive unit according to the moving path for autonomous movement.

The invention discloses an RGB-D image object detection and semantic segmentation method based on a deep convolution network, which belongs to the field of depth learning and machine vision. According to the method provided by the technical scheme of the invention, Faster-RCNN is used to replace the original slow RCNN; Faster-RCNN uses GPU, which is fast in the aspect of feature extracting, and at the same time generates a regional scheme in the network; the whole training process is training from end to end; FCN is used to carry out RGB-D image semantic segmentation; FCN uses a GPU and the deep convolution network to rapidly extract the deep features of an image; deconvolution is used to fuse deep features and shallow features of the image convolution; and the local semantic information of the image is integrated into the global semantic information.

An interruption free navigator includes an inertial measurement unit, a north finder, a velocity producer, a positioning assistant, a navigation processor, an altitude measurement, an object detection system, a wireless communication device, and a display device and map database. Output signals of the inertial measurement unit, the velocity producer, the positioning assistant, the altitude measurement, the object detection system, and the north finder are processed to obtain highly accurate position measurements of the person. The user's position information can be exchanged with other users through the wireless communication device, and the location and surrounding information can be displayed on the display device by accessing a map database with the person position information.

An ergonomic water conserving faucet assembly that pivots around a cognitive central point providing touchless water temperature, flow rate, volume control and spray pattern adjustment through multiple, hygienic means. The assembly comprises a pivotable, ergonomic, ball-shaped spout that may be used statically or hand held; a retractable water delivery hose connecting the spout to a water source; a water mixing valve at the water source delivering water of preselected temperature; solenoid valves controlling flow; proximity and object detection sensors mapping the sink area and detecting input signals; speech sensors with microphone for voice control; an LED display of water temperature; internal speakers delivering audible prompts; and an electronic controller recognizing speech and supervising operations.

In an image capturing apparatus, a video input unit (2) captures the image of an object and sequentially acquires image data associated with the image capturing, a model data memory (6) stores model data associated with the first feature quantity calculated from a feature point of the object in a model image, a principal object detection unit (3) calculates the second feature quantity from a feature point of the object in the acquired image data, a state change estimation unit (4) estimates, on the basis of the second feature quantity and the model data, the timing when the object satisfies a predetermined condition, and an image input processing control unit (7) stores the image data corresponding to the estimated timing in an image recording unit (5). This configuration makes the image capturing apparatus acquire an image in a more proper state without large-capacity memory.

Provided is a system and method for object detection and tracking in a video stream. Frames of the video stream are divided into regions of interest and a probability that the region contains at least a portion of an object to be tracked is calculated for each region of interest. The regions of interest in each frame are then classified based on the calculated probabilities. A region of interest (RI) frame is then constructed for each video frame that reports the classification of regions of interest in the video frame. Two or more RI frames are then compared in order to determine a motion of the object. Also provided is a system executing the presently described method, as well as a device including the system. The device may be for example, a portable computer, a mobile telephone, or an entertainment device.

An object recognizing means obtains a relative physical value between a movable body and an object such as at least one of the other movable body, an object on the ground, a position on the ground, a topographical feature and a regional information around the movable body, and a safety keeping area calculator calculates an imaginary safety keeping area around the movable body from the relative physical value, and an object intrusion judging means decides as to at least one of whether or not the object is within the safety keeping area and whether or not the object will be within the safety keeping area to perform at least one of controlling for preventing a collision between the movable body and the object and outputting an alarm, when deciding the at least one of that the object is within the safety keeping area and that the object will be within the safety keeping area.

Methods of detecting an object in an image using a convolutional neural-network-based architecture that processes multiple feature maps of differing scales from differing convolution layers within a convolutional network to create a regional-proposal bounding box. The bounding box is projected back to the feature maps of the individual convolution layers to obtain a set of regions of interest (ROIs) and a corresponding set of context regions that provide additional context for the ROIs. These ROIs and context regions are processed to create a confidence score representing a confidence that the object detected in the bounding box is the desired object. These processes allow the method to utilize deep features encoded in both the global and the local representation for object regions, allowing the method to robustly deal with challenges in the problem of object detection. Software for executing the disclosed methods within an object-detection system is also disclosed.

Provided herein are methods and systems for implementing three-dimensional perception in an autonomous robotic system comprising an end-to-end neural network architecture that directly consumes large-scale raw sparse point cloud data and performs such tasks as object localization, boundary estimation, object classification, and segmentation of individual shapes or fused complete point cloud shapes.

A material handling system comprising: a platform for receiving a plurality of articles; a plurality of sorting bins, each bin in corresponding to an article type; an object recognition system for identifying an article on the platform and categorizing said article by specific article type; said 3D object recognition system including a 3D detector for detecting said article and a database containing data on characteristics of article types; an article engagement assembly for engaging and delivering said article to said corresponding sorting bin; wherein the object recognition system is arranged to identify the article based upon the data within said database and instructs the article engagement assembly to engage and place the article in said sorting bin.

A light sensor system includes at least one light emitter, a light sensor unit and a processing unit. The light sensor unit is arranged to receive reflected light from an object in accordance with a time sequence in which the at least one light emitter is activated, and accordingly output a plurality of reflected signals. The processing unit is arranged to receive the reflected signals, identify a signal function of time by referring to occurrence sequence of local peak levels of the reflected signals, and determine motion of the object according to the signal function of time. Another light sensor system is proposed. The major difference between the two light sensor systems is that the processing unit of the another light sensor system is arranged to identify the signal function of time by comparing a predetermined threshold with signal levels of the reflected signals.

A monitor camera photographs a whole monitor area for capturing the photographed whole monitor area as high definition image data higher in resolution than a display screen. A new object detection/extraction unit detects a new object in the monitor area based on the high definition image data captured by the monitor camera, extracts partial image data of an area, which contains the new object, from the high definition image data, and obtains location information on the partial image data in relation to the high definition image data. A whole-monitor-area image data down-sampling unit down-samples the high definition image data to produce standard definition image data corresponding to the resolution of the display screen. An image enlargement unit enlarges the standard definition image data based on entered enlargement instruction information. An image combination unit overlaps the partial image data on image data, enlarged by the image enlargement unit, based on the location information and sends resulting image data to the display screen.

An object awareness determination system and method of determining awareness of a driver of a vehicle to an object is provided. The system includes an object monitor including an object detection sensor for sensing an object in a field of view and determining a position of the object. The system also includes an eye gaze monitor including an imaging camera oriented to capture images of the vehicle driver including an eye of the driver. The gaze monitor determines an eye gaze vector. The system further has a controller for determining driver awareness of the object based on the detected object position and the eye gaze vector.