首页>
外国专利>
METHOD FOR CLASSIFYING ACCIDENT VIDEO BY SPEED PROFILE ANALYSIS
METHOD FOR CLASSIFYING ACCIDENT VIDEO BY SPEED PROFILE ANALYSIS
展开▼
机译:通过速度剖面分析对事故视频进行分类方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
The present invention selects an actual accident image by using a speed profile analysis before and after an impact event, and deletes or converts the image with a grade determined to have a low probability of an accident from a storage medium, or transfers it to a cloud server. The present invention relates to a method for selecting an accidental video by analyzing a speed profile to ensure that the usable capacity of the storage medium is sufficiently secured, the amount of data transmission and the cost can be reduced, and the loss of unnecessary management costs can be prevented by taking measures to be prohibited. In the method for selecting an accident image by analyzing a speed profile of the present invention, after detecting an impact applied to the vehicle by a black box installed in the vehicle, the image before and after the impact detection time (T0) is stored as an impact detection image. In the accident image selection method for selecting whether or not this accident image, (a) the driving speed of the vehicle is in a section between the first time point T1 before the impact detection time point T0 and the second time point T2 after the impact detection time point T0 Determining whether or not the reference speed (V TH) or less is maintained for a predetermined time; (b) if it is determined in step (a) that the driving speed exceeds the reference speed (V TH ), determining the shock detection image as an image with a low probability of an accident; (c) determining the impact detection image as an image with a high probability of an accident when the driving speed is determined to be less than or equal to the reference speed (V TH ) in step (a); (d) A deceleration ratio of the vehicle is detected in a section between the first time point T1 and the second time point T2, and i) the deceleration rate is a predetermined safety deceleration rate (SDR: Safety Deceleration Ratio), ii) whether the deceleration rate is greater than the safe deceleration rate, and is less than or equal to a predetermined accident deceleration ratio (ADR) as a value larger than the safe deceleration rate, iii) the deceleration rate is the Determining whether the accident deceleration rate is exceeded; (e) as a result of the determination in step (d), i) if the deceleration rate is less than the safe deceleration rate, moving the impact detection image to an area that can be deleted or overwritten from the storage medium of the black box; (f) As a result of the determination in step (d), ii) when the deceleration rate is greater than the safe deceleration rate and less than the accident deceleration rate, the shock detection image is temporarily stored in the storage medium of the black box and remotely controlled. Transmitting to the control server when there is a request for transmission from the server; And (g) when the determination result of step (d), iii) the deceleration rate exceeds the accident deceleration rate, determining the impact detection image as an accident image. According to the present invention, the vehicle accident image at low speed is accurately converted to the actual accident image by analyzing the speed profile before and after the impact event without the need to adjust the shock detection sensitivity. It is possible to select, and by deleting the impact detection image determined to have a low probability of an accident or moving to an overwriteable area, there is an effect of remarkably reducing the loss cost required for selecting an accident image.
展开▼