Techniques for automatic image tagging and selection at a mobile device include generating a smart image tagging model by first training an initial model based on different angles of image capture of subject vehicles, and then re-training the trained model using weights discovered from the first training and images that have been labeled with additional tags indicative of different vehicle portions and/or vehicle parameters. Nodes that are training-specific are removed from the re-trained model, and the lightweight model is serialized to generate the smart image tagging model. The generated model may autonomously execute at an imaging device to predict respective tags associated with a stream of frames; select, capture and store respective suitable frames as representative images corresponding to the predicted tags; and provide the set of representative images and associated tags for use in determining vehicle damage, insurance claims, and the like.

A computer-implemented method in which one or more processing devices perform operations may include obtaining a field image of a railcar collected from a field camera system and applying a machine-learning algorithm to the field image to generate a machine-learning algorithm output. The method may also include performing a post-processing operation on the machine-learning algorithm output to generate a filtered machine-learning algorithm output. Further, the method may include detecting a defect of the railcar using the filtered machine-learning algorithm output.

Methods and systems for facilitating photo-based estimation are described. In an aspect, a computing device is configured to receive, from the camera, a signal comprising image data. The image data may represent at least a portion of a vehicle. The computing device may also retrieve data representing a preferred scene of the vehicle and determine, based on the image data and based on the data representing the preferred scene of the vehicle, whether the image data corresponds to the preferred scene of the vehicle. The server may, when the received image data corresponds to the preferred scene of the vehicle, enable capture of the image data. The server may send, via the communications module, a signal representing the captured image data to a processing server configured to analyze the captured image data to assess vehicular damage.

A method that is able to measure the gap and flush of vehicle parts by means of a measuring tunnel. The method is able to determine the coordinates in 3D of the edges or ends of two adjacent parts of a vehicle. The measuring tunnel includes several video cameras, LED lights, a conveyor, a position encoder that measures the movement of the vehicle; a total station that measures fixed points of the measuring tunnel structure; a calibration chessboard and a calibration pattern; processing and storage means to store images taken by the video cameras, Computer-Aided-Design files of the vehicles and an edge recognition algorithm.

A computer-implemented method in which one or more processing devices perform operations may include obtaining a field image of a railcar collected from a field camera system and applying a machine-learning algorithm to the field image to generate a machine-learning algorithm output. The method may also include performing a post-processing operation on the machine-learning algorithm output to generate a filtered machine-learning algorithm output. Further, the method may include detecting a defect of the railcar using the filtered machine-learning algorithm output.

Even in a case where an external camera image-captures a door of a car, with high precision, it can be set to be determined whether or not a foreign substance inserted into the door is present. A monitoring apparatus acquires a camera image that results from a camera image-capturing the vicinity of a door of a car using an external camera, from a camera. A processor of the monitoring apparatus detects the door from the camera image, sets a determination area based on a position of the detected door, acquires an image of the determination area from the camera image, determines whether or not a foreign substance inserted into the door is present, based on the image of the determination area, and performs report outputting to a monitor and a reporting apparatus according to a result of the determination.

The invention provides a verification method of the dynamic virtual image display distance of a user interface, comprising the following steps: creating a tested image database; wherein the tested image database comprises a tested image displayed according to a standard virtual image display distance; displaying a first tested image; projecting a first image on a stacked image element; wherein the first image is displayed at a first virtual image display distance, which is the same with a first standard virtual image display distance of the first tested image; capturing the first tested image and the first image; performing a first reliability evaluation procedure for the first image and the first tested image; and calculating a first overlap ratio for the first image and the first tested image to verify accuracy of the user interface.

A method of inspecting a vehicle includes: acquiring a to-be-inspected image of an inspected vehicle (S11); acquiring a visual feature of the to-be-inspected image using a first neural network model (S12); retrieving a template image from a vehicle template library based on the visual feature of the to-be-inspected image (S13); determining a variation region between the to-be-inspected image and the template image (S14); and presenting the variation region to a user (S15). The system of inspecting a vehicle includes a radiation imaging device (150), a display device (130), an image processor (140), and a storage device (120). The present disclosure further includes a computer-readable storage medium.

A vehicle analysis environment includes one or more display screens, such as a display screen wall or an array of display screens. While a vehicle is in the vehicle analysis environment, a vehicle analysis system renders and displays one or more vehicle sensor calibration targets and/or one or more simulated events on the one or more display screens. Vehicle sensors of the vehicle capture sensor data while in the vehicle analysis environment. The sensor data depict the vehicle sensor calibration targets and/or the simulated events that are displayed on the one or more display screens. The vehicle can output actions based on the simulated event and/or can calibrate its vehicle sensors based on the vehicle sensor calibration targets.

A non-transitory computer-readable medium and methods capable of being executed thereby are provided. The non-transitory computer-readable medium includes contents that are configured to cause a computing system to perform the method related to a charging port of an electrified vehicle. The method, or methods, include taking a current photo of at least one of a charging receptacle or a cordset of the charging port; image processing the current photo; and determining whether the current photo includes one or more identified faults from the image processed current photo. If one or more identified faults are present in the image processed current photo, at least one of: sending a maintenance message; limiting charging; or adjusting system parameters.