A system for updating training data includes an interface and a processor. The interface is configured to receive a set of vehicle data. The set of vehicle data includes images and assigned labels associated with the images. The processor is configured to determine a set of training data and a set of test data from the set of vehicle data; train a model with the set of training data; determine a set of predicted labels for the set of vehicle data using the model; identify a set of potential mislabeled data using the set of predicted labels and the assigned labels; and determine an updated set of training data by relabeling the set of potential mislabeled data and replacing the set of potential mislabeled data with a relabeled set of data.

A center information display system includes a center information display; a capturing device configured to obtain information on an occupant inside a vehicle; a rotating structure configured to adjust a view angle of the center information display; and a control device in communication with the capturing device, being configured to determine whether there is an occupant in a driver seat or a passenger seat, and control the rotating structure to rotate the center information display to a predetermined view angle.

Systems and methods for using image analysis techniques to assess abnormal vehicle operating conditions are disclosed. According to aspects, a computing device may access and analyze image data depicting an individual(s) within a vehicle. Based on the depicted individuals(s) and optionally on other data, the computing device may determine that an abnormal condition exists. In response, the computing device may generate a notification and transmit the notification to an electronic device of an individual associated with the vehicle.

A safe exiting assistance system includes an object detector that detects an object which approaches from a rear side of a vehicle, a body detector that detects physical information of a passenger in the vehicle, a door opening/closing detector that detects door opening or closing information of the vehicle, a determiner that determines whether a situation requires alert to the passenger in response to detection results of the object detector and the door opening/closing detector and calculates a value sensed by the body detector in response to vehicle information set in advance, and a controller that differently controls a position and an irradiation angle of a light source in response to the physical information when information in which the alert is necessary is received from the determiner.

According to exemplary methods of training a convolutional neural network, input images are received into a computerized device having an image processor. The image processor evaluates the input images using first convolutional layers. The number of first convolutional layers is based on a first size for the input images. Each layer of the first convolutional layers receives layer input signals comprising features of the input images and generates layer output signals that include signals from the input images and ones of the layer output signals from previous layers within the first convolutional layers. Responsive to an input image being a second size larger than the first size, additional convolutional layers are added to the convolutional neural network. The number of additional convolutional layers is based on the second size in relation to the first size. The additional convolutional layers are initialized using weights from the first convolutional layers. Feature maps comprising the layer output signals are created.

The present disclosure is directed to apparatuses, systems and methods for automatically classifying images of occupants inside a vehicle. More particularly, the present disclosure is directed to apparatuses, systems and methods for automatically classifying images of occupants inside a vehicle by comparing current image feature data to previously classified image features.

An amusement park attraction system includes a plurality of autonomous vehicles configured to accommodate one or more guests and located in an attraction dispatch area. An individual autonomous vehicle includes a vehicle controller including a memory storing instructions and a processor configured to execute the instructions, the instructions configured to cause the vehicle controller to receive an indication that the individual autonomous vehicle is loaded at capacity within the dispatch area and drive autonomous operation of the individual autonomous vehicle from the dispatch area to enter the ride path.

Provided is a technique for selecting a cancellation filter coefficient that can stably cancel noise for a user using a seat of an aircraft, an automobile, or the like. A cancellation filter coefficient selection apparatus includes: a recording unit configured to record cancellation filter coefficients; a seat position acquisition unit configured to acquire a position of a seat used by a user; and a cancellation filter coefficient selection unit configured to, from among the cancellation filter coefficients recorded in the recording unit, select a cancellation filter coefficient corresponding to the position of the seat as a cancellation filter coefficient to be used in filtering for generating a cancellation signal for cancelling noise at a position where quiet is desired, based on a reference signal output by a reference microphone configured to collect noise in a predetermined space. The reference microphone and a speaker for emitting sound based on the cancellation signal are installed on the seat, and positions of the reference microphone and the speaker in the predetermined space change accompanying movement of the seat.

An imaging system is provided having an image sensor, an electrowetting lens positioned in front of the image sensor and configured to change at least one optical characteristic in response to an electrical stimulus so as to change a field of view of the image sensor, and a controller coupled to the electrowetting lens and configured to select a field of view of the image sensor by selecting the electrical stimulus to be applied to the electrowetting lens.

A system and method for detecting seatbelt positioning includes capturing, by a camera, a near infrared source image of an occupant in a vehicle, determining a position of a head of the occupant in the source image, determining a region of interest for a seatbelt based on the position of the head; determining an oversaturation condition in the region of interest for the seatbelt, adjusting the source image to form an adjusted image compensating the oversaturation condition in the region of interest for the seatbelt, converting the adjusted image to a black-and-white image, and detecting the seatbelt within the black-and-white image. Adjusting the source image to compensate the oversaturation condition may include adjusting an exposure level of the source image based on a contrast ratio in the region of interest and/or based on detecting a predetermined pattern of the seatbelt within the region of interest for the seatbelt.