An electronic device that includes a vision system carried by a bracket assembly is disclosed. The vision system may include a first camera module that captures an image of an object, a light emitting element that emits light rays toward the object, and a second camera module that receives light rays reflected from the object. The light rays may include infrared light rays. The bracket assembly is designed not only carry the aforementioned modules, but to also maintain a predetermined and fixed separation between the modules. The bracket assembly may form a rigid, multi-piece bracket assembly to prevent bending, thereby maintaining the predetermined separation. The electronic device may include a transparent cover designed to couple with a housing. The transparent cover includes an alignment module designed to engage a module and provide a moving force that aligns the bracket assembly and the modules to a desired location in the housing.

A non-contact method of characterizing the isostatic strength of a ceramic member or article includes capturing a digital image of the ceramic article, and then forming a two-dimensional representation of the ceramic article and the web therein based on the captured digital image. The method also includes performing finite-element analysis on the two-dimensional representation of the ceramic article using a select amount of simulated isostatic pressure to determine a maximum stress value within the two-dimensional representation of the web. The method further includes using the maximum stress value to characterize the isostatic strength of the ceramic article.

In one embodiment, a method or system generates a high resolution 3-D point cloud to operate an autonomous driving vehicle (ADV) from a low resolution 3-D point cloud and camera-captured image(s). The system receives a first image captured by a camera for a driving environment. The system receives a second image representing a first depth map of a first point cloud corresponding to the driving environment. The system upsamples the second image by a predetermined scale factor to match an image scale of the first image. The system generates a second depth map by applying a convolutional neural network (CNN) model to the first image and the upsampled second image, the second depth map having a higher resolution than the first depth map such that the second depth map represents a second point cloud perceiving the driving environment surrounding the ADV.

A method and apparatus for controlling an image acquisition device are provided. The method includes that: multiple images photographed by multiple image acquisition devices are acquired; an active image acquisition device and an inactive image acquisition device in the multiple image acquisition devices are determined according to the multiple images; the active image acquisition device is controlled to photograph a first image by using a first configuration, and the inactive image acquisition device is controlled to photograph a second image by using a second configuration; the first image transmitted by the active image acquisition device is acquired, and the second image transmitted by the inactive image acquisition device is acquired, and a bandwidth required by the active image acquisition device to transmit the first image is greater than a bandwidth required by the inactive image acquisition device to transmit the second image.

A method and apparatus for controlling an image acquisition device are provided. The method includes that: multiple images photographed by multiple image acquisition devices are acquired; an active image acquisition device and an inactive image acquisition device in the multiple image acquisition devices are determined according to the multiple images; the active image acquisition device is controlled to photograph a first image by using a first configuration, and the inactive image acquisition device is controlled to photograph a second image by using a second configuration; the first image transmitted by the active image acquisition device is acquired, and the second image transmitted by the inactive image acquisition device is acquired, and a bandwidth required by the active image acquisition device to transmit the first image is greater than a bandwidth required by the inactive image acquisition device to transmit the second image.

A railroad track inspection system has multiple track scanning sensors, a data store, and a scan data processor. The scan data processor provides automatic analysis of the track scan data to detect track components within the scan data from a predetermined list of component types according to features identified in said scan data. The track scanning sensors, data store and scan data processor are attached to a common support structure for mounting the system to a railway vehicle in use. An inertia sensor and common master clock are used to make corrections to the output of the track scanning sensors to accommodate dynamic forces in use. The inspection system may be provided in a single housing for mounting to a conventional passenger/freight rail vehicles and may operate automatically in an unattended mode. The location of track components and/or defects may be logged.

The invention relates to processing and generating image data and analysis and displaying a 3D image. A method of controlling a device for generating an augmented reality consists in receiving object-related data from sensors, recognizing the object and generating a virtual object, the authenticity of an activation code that identifies a set of pictures containing objects is pre-verified, data related to an object obtained in the form of photo or video frame(s), the object is identified on said frame(s) by comparison with images stored in the memory of the user device, and a virtual object is generated in the form of a 3D model, reproduced in real time, on a display of the user device, on top of the acquired frames. Wherein the object is a two-dimensional image of at least one item depicted on a picture contained in a set to which an activation code is assigned.

In a fingerprint sensor, two adjacent sensing plates are detected at a same time to obtain first sensing data and second sensing data therefrom respectively, so that the first sensing data and the second sensing data include a same noise, then the first sensing data is subtracted from the second sensing data to generate a difference by which the noise is eliminated, and the difference is added to first fingerprint data corresponding to the sensing plate which provides the first sensing data, resulting in second fingerprint data corresponding to the other sensing plate. The fingerprint portions on the two adjacent sensing plates are determined according to the first and the second fingerprint data respectively.

A method and circuit for fingerprint detection perform a first round of edge detection with a first parameter, and if a first value obtained during the first round of edge detection does not satisfy a predetermined threshold, perform a second round of edge detection with a second parameter, so as to allow a fingerprint that is not acceptable for the first parameter to be accepted for fingerprint recognition. If a second value obtained during the second round of edge detection does not satisfy the predetermined threshold neither, it is determined that the object subject to fingerprint detection is not a finger.

An image processing apparatus includes a first memory used for first rearrangement processing on a group of pixels in an input image, and a second memory used for second rearrangement processing on a group of pixels in an image obtained by the first rearrangement processing, and performs correction processing that includes the first rearrangement processing and the second rearrangement processing on the input image. One of the first and second memories is capable of higher-speed random access than the other memory and has a smaller memory capacity than the other memory. One of the first rearrangement processing and the second rearrangement processing is processing for rearranging a group of pixels in each of a plurality of block images generated from the input image, and the other rearrangement processing is processing for rearranging pixel rows among the block images. The one rearrangement processing involves random access to the one memory.