6 results about "Pedestrian detection" patented technology

The invention discloses a method for constructing a pedestrian detection model, and belongs to the technical field of graphic processing. The method for constructing the pedestrian detection model comprises the following steps: randomly selecting a pedestrian image from a database, marking the head of the pedestrian image as a label file, and dividing the selected pedestrian image into a trainingset, a test set and a verification set; performing clustering analysis on the training set to obtain a new prior frame, and performing network structure adjustment on the YOLOv3 network; and trainingthe training set by using a YOLOv3 network to obtain a pedestrian detection model, evaluating the pedestrian detection model by using a verification set, and finally testing the pedestrian detection model by using a test set. The invention aims to overcome the defect that in the prior art, when pedestrians in a monitoring scene are in a dense state, the pedestrian detection result is inaccurate, and provides the method for constructing the pedestrian detection model, so that the problem of pedestrian shielding during pedestrian detection can be solved, and the pedestrian detection accuracy isimproved.

The embodiment of the invention provides a pedestrian detection method and device. The pedestrian detection method comprises the steps that a to-be-processed image is acquired; scene information of each scene to which pixels belong in the to-be-processed image is analyzed; and pedestrians in the to-be-processed image are detected in combination with the scene information of each scene to which the pixels belong so as to determine locations of the pedestrians in the to-be-processed image. According to the pedestrian detection method and device, pedestrian detection is performed in combination with the scene information in the image, false positive results generated by a pedestrian detection algorithm can be effectively reduced by use of the scene information, and meanwhile the detection precision of the pedestrian detection algorithm can be improved by use of the scene information.

Methods and apparatus are provided for improved abandoned object recognition using pedestrian detection. An abandoned object is detected in one or more images by determining if one or more detected objects in a foreground of the images comprises a potential abandoned object; applying a trained pedestrian detector to the potential abandoned object to determine if the potential abandoned object comprises at least a portion of a pedestrian; and classifying the potential abandoned object as an abandoned object based on whether the potential abandoned object is not at least a portion of a pedestrian. The trained pedestrian detector is trained using positive training samples comprised of at least portions of human bodies in one or more poses and/or negative training samples comprised of at least portions of abandoned objects.

A camera-based modular electronic exterior mirror system includes a camera assembly and a display assembly, the camera assembly comprises a camera module, a camera support and a camera wire harness. The display assembly comprises a display screen and an image processing unit. The camera module installed in the camera support is connected with the image processing unit through the camera wire harness. The display screen is arranged on the driver side A-pillar or on the right side area of the steering wheel. In application, the camera module transfers the road state image taken by the camera module into the image processing unit through the camera wire harness, and the image processing unit processes the received image to obtain narrow-angle and wide-angle views, and arranges the images up and down on the display screen, and can automatically judge whether there is a foreign object in the blind area through the image recognition software, and an alarm is sent when there has the foreign object. This design not only has good quality of image display and strong effect of rear-view guidance, but also has pedestrian detection function and a wide range of adaptability.