Apparatuses, methods, and systems are presented for reacting to scene-based occurrences. Such an apparatus may comprise dedicated computer vision (CV) computation hardware configured to receive sensor data from a sensor array comprising a plurality of sensor pixels and capable of computing one or more CV features using readings from neighboring sensor pixels of the sensor array. The apparatus may further comprise a first processing unit configured to control operation of the dedicated CV computation hardware. The first processing unit may be further configured to execute one or more application programs and, in conjunction with execution of the one or more application programs, communicate with at least one input/output (I/O) device controller, to effectuate an I/O operation in reaction to an event generated based on operations performed on the one or more computed CV features.

An aerial robot includes an image sensor for capturing images of an environment. The robot receives a first image captured at a first location. The robot identifies one or more first pixels in the first image. The first pixels correspond to one or more targeted features of an object identified in the first image. The robot receives a second image captured at the second location. The robot receives its distance data that estimates a movement of the robot from the first location to the second location. The robot identifies second pixels in the second image. The second pixels corresponding to the targeted features of the object as appeared in the second image. The robot determines an estimated distance between the robot and the object based on the changes of locations of the second pixels from the first pixels relative to the movement of the robot provided by the distance data.

Techniques for accurately predicting and avoiding collisions with objects detected in an environment of a vehicle are discussed herein. A vehicle safety system can implement one or more models to output data indicating an intersection probability between the object and a portion of the vehicle in the future. The model may employ a blur filter to modify a representation of an object usable to determine the intersection probability. A vehicle can be controlled in an environment based at least in part on the intersection probability.

Apparatuses, methods, and systems are presented for sensing scene-based occurrences. Such an apparatus may comprise a vision sensor system comprising a first processing unit and dedicated computer vision (CV) computation hardware configured to receive sensor data from at least one sensor array comprising a plurality of sensor pixels and capable of computing one or more CV features using readings from neighboring sensor pixels. The vision sensor system may be configured to send an event to be received by a second processing unit in response to processing of the one or more computed CV features by the first processing unit. The event may indicate possible presence of one or more irises within a scene.

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.

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.

This disclosure relates generally to ge visual inspection systems, and more particularly to a method and system for facilitating real time detection of linear infrastructural objects in aerial imagery. In one embodiment, a background suppression technique is applied to one or more hardware processors to a HSV image. Further, a mean shift filtering technique is applied to the hardware processors to find a peak of a confidence map and then a gradient image generation is performed for a plurality of edges of the image. A seed point pair along a middle cut portion of a linear feature of the HSV image to identify one or more boundaries of the seed point pair is extracted and then a contour growing approach to detect the boundaries of the linear feature is initiated. Lastly, one or more false positives are removed by using a rigidity feature, the rigidity feature being equivalent to the total sum of gradient orientations.

A method for recognizing different object-categories within images based on texture classification of the different categories, which is implemented in an electronic device, includes extracting texture features from block images segmented from original images according to at least one Gabor filter; determining a grayscale level co-occurrence matrix of each block image according to the texture features; calculating texture feature statistics of each block image according to the grayscale level co-occurrence matrix; training and generating an object recognition model using the texture features and the texture feature statistics; and recognizing and classifying at least one object in original image according to the object recognition model.

Disclosed is an image analysis method including inputting analysis data, including information regarding an analysis target cell to a deep learning algorithm having a neural network structure, and analyzing an image by calculating, by use of the deep learning algorithm, a probability that the analysis target cell belongs to each of morphology classifications of a plurality of cells belonging to a predetermined cell group.

An aerial robot includes an image sensor for capturing images of an environment. The robot receives a first image captured at a first location. The robot identifies one or more first pixels in the first image. The first pixels correspond to one or more targeted features of an object identified in the first image. The robot receives a second image captured at the second location. The robot receives its distance data that estimates a movement of the robot from the first location to the second location. The robot identifies second pixels in the second image. The second pixels corresponding to the targeted features of the object as appeared in the second image. The robot determines an estimated distance between the robot and the object based on the changes of locations of the second pixels from the first pixels relative to the movement of the robot provided by the distance data.