Search

CN-122009026-A - Rearview mirror display method, vehicle and electronic equipment

CN122009026ACN 122009026 ACN122009026 ACN 122009026ACN-122009026-A

Abstract

The application relates to a rearview mirror display method, a vehicle and electronic equipment, which belong to the technical field of image processing, and comprise the steps of acquiring an infrared image and a visible light image, carrying out spatial registration on the infrared image based on the visible light image, carrying out edge detection on the registered infrared image, identifying to obtain at least one target graph object, carrying out comprehensive evaluation on the target graph object based on position information, size information and heat radiation intensity information of the target graph object in the infrared image to obtain a comprehensive evaluation value, comparing the comprehensive evaluation value with a preset display threshold, and carrying out enhanced display on the target graph object with the comprehensive evaluation value larger than or equal to the preset display threshold in a display picture of the rearview mirror.

Inventors

  • DONG CHAOQI
  • JIA XIONG

Assignees

  • 长城汽车股份有限公司

Dates

Publication Date
20260512
Application Date
20260130

Claims (10)

  1. 1. A rearview mirror display method, characterized in that it comprises: Acquiring an infrared image and a visible light image, and performing spatial registration on the infrared image based on the visible light image; performing edge detection on the registered infrared image, and identifying to obtain at least one target graphic object; Based on the position information, the size information and the heat radiation intensity information of the target graphic object in the infrared image, comprehensively evaluating the target graphic object to obtain a comprehensive evaluation value; and comparing the comprehensive evaluation value with a preset display threshold value, and performing enhanced display on a display picture of the rearview mirror for target graphic objects of which the comprehensive evaluation value is larger than or equal to the preset display threshold value.
  2. 2. The rearview mirror display method according to claim 1, wherein the comprehensively evaluating the target graphic object based on the position information, the size information, and the heat radiation intensity information of the target graphic object in the infrared image includes: calculating to obtain a position score of the target graphic object based on the centroid coordinates of the target graphic object; Calculating to obtain an area score of the target graphic object based on the area proportion of the target graphic object in the infrared image; Calculating to obtain a temperature score of the target graphic object based on a normalized ratio of the average gray value of the target graphic object graphic area to the whole gray range of the infrared image; And calculating and obtaining a comprehensive evaluation value of the target graphic object based on the position score, the area score and the temperature score.
  3. 3. The rearview mirror display method according to claim 2, wherein calculating a position score of the target graphic object based on centroid coordinates of the target graphic object includes: Obtaining centroid coordinates of the target graphic object; Establishing a normalized coordinate system taking the midpoint of the lower edge of the infrared image as an origin, and converting the centroid coordinates into the normalized coordinate system; Calculating a position score of the target graphic object by a predetermined position scoring function based on the converted coordinates; wherein the position scoring function is configured such that the score value decreases as the coordinate value of the coordinate in the height direction increases and decreases as the coordinate deviates from the origin in the horizontal direction.
  4. 4. The rearview mirror display method according to claim 2, wherein calculating an area score of the graphic object based on an area ratio of the graphic object in the infrared image includes: Acquiring a first pixel number of the target graphic object in the infrared image; determining the area proportion of the target graphic object in the infrared image based on the ratio of the first pixel number to the total pixel number of the infrared image; the area ratio is mapped to the area score based on a logarithmic operation relationship.
  5. 5. The rearview mirror display method according to claim 2, wherein calculating the temperature score of the graphic object based on the normalized ratio of the average gray value of the graphic object graphic area to the entire gray range of the infrared image comprises: Obtaining a maximum gray value and a minimum gray value of the infrared image; calculating the average gray value of the graph area corresponding to the target graph object; carrying out normalization calculation based on the average gray value of the target graphic object and the maximum gray value and the minimum gray value of the infrared image to obtain a normalization ratio; mapping the normalized ratio to the temperature score by a predetermined linear mapping function.
  6. 6. The rearview mirror display method of any one of claims 1-5, wherein said edge detection of said registered infrared images, identifying at least one target graphical object, comprises: performing edge detection on the infrared image to obtain an edge image; Performing Hough transformation on the edge image, identifying edge line segments, and connecting line segments which are mutually adjacent and have similar directions to form a closed graph outline; And carrying out connected domain analysis on the closed graph outline, and marking different connected domains as different target graph objects.
  7. 7. The rearview mirror display method of any one of claims 1-5, wherein the acquiring an infrared image and a visible light image and spatially registering the infrared image based on the visible light image comprises: Acquiring the infrared image originally acquired by an infrared camera; Acquiring the visible light image originally acquired by a visible light camera; And translating the infrared image based on a pre-calibrated translation parameter to spatially align the infrared image with the visible light image, thereby obtaining the registered infrared image.
  8. 8. The rearview mirror display method according to any one of claims 1 to 5, characterized in that the performing enhancement display in a display screen of the rearview mirror includes: And for the target graphic object with the comprehensive evaluation value being greater than or equal to the preset display threshold value, performing superposition display in a display picture of the rearview mirror in a form of a highlight contour line.
  9. 9. A vehicle comprising a controller for implementing the rearview mirror display method according to any one of claims 1 to 8.
  10. 10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of displaying a rear view mirror according to any one of claims 1 to 8 when executing the computer program.

Description

Rearview mirror display method, vehicle and electronic equipment Technical Field The present application relates to the field of image processing technologies, and in particular, to a rearview mirror display method, a vehicle, and an electronic device. Background With the rapid development of automobile electronic technology, streaming media rearview mirrors have gradually become an important configuration for improving driving safety. Compared with the traditional optical internal rearview mirror, the streaming media rearview mirror collects real-time video through the vehicle tail camera and displays the video on the display screen in the vehicle, effectively eliminates the shielding of structures such as objects, rear passengers or C columns in the carriage to the view field, and provides a clearer tail view field. However, even if the camera adopted by the streaming media rearview mirror has a certain night vision function in low-illumination or severe weather environments such as night, tunnels, foggy days, etc., the recognition capability of the streaming media rearview mirror on potential dangerous targets still has a limitation, and the targets may be difficult to be found in time by a driver due to insufficient ambient light and low contrast with the background. In the prior art, the infrared image and the visible light image are generally identified in a one-to-one correspondence manner, and because of the complex road information, the edge features identified by the infrared image also have a lot of messy information, such as road surface heat residues, remote non-road area heat sources and the like. After the simple correspondence, a large number of non-critical and non-threatening heat source outlines are simultaneously displayed on the streaming media rearview mirror images, so that not only the targets with the real safety threat cannot be effectively highlighted, but also the information overload and the visual interference are caused, the attention of a driver is dispersed, and the driving safety is reduced. Disclosure of Invention The present application solves at least one of the technical problems in the related art to a certain extent. To this end, the present application aims to provide a rearview mirror display method, a vehicle and an electronic device. To achieve the above object, the present application provides a rearview mirror display method, comprising: Acquiring an infrared image and a visible light image, and performing spatial registration on the infrared image based on the visible light image; performing edge detection on the registered infrared image, and identifying to obtain at least one target graphic object; Based on the position information, the size information and the heat radiation intensity information of the target graphic object in the infrared image, comprehensively evaluating the target graphic object to obtain a comprehensive evaluation value; and comparing the comprehensive evaluation value with a preset display threshold value, and performing enhanced display on a display picture of the rearview mirror for target graphic objects of which the comprehensive evaluation value is larger than or equal to the preset display threshold value. According to the technical scheme, the multi-dimensional comprehensive evaluation is performed by combining the position, the size and the heat radiation intensity of the target, so that the high threat target and the low threat interference source can be distinguished, the accuracy of dangerous target identification is remarkably improved, false alarm is effectively reduced, on the basis, the preset threshold is utilized for screening and only the high risk target is enhanced and displayed, information overload is avoided, visual presentation is optimized, a driver is guided to pay attention to the key threat rapidly, the cognitive load of the driver is reduced, various severe environments such as night, foggy days and the like can be reliably handled, and the practicability and the safety auxiliary efficiency of the rearview mirror under a complex scene are enhanced. In some embodiments of the present application, the performing the comprehensive evaluation on the target graphic object based on the position information, the size information and the thermal radiation intensity information of the target graphic object in the infrared image to obtain the comprehensive evaluation value includes: calculating to obtain a position score of the target graphic object based on the centroid coordinates of the target graphic object; Calculating to obtain an area score of the target graphic object based on the area proportion of the target graphic object in the infrared image; Calculating to obtain a temperature score of the target graphic object based on a normalized ratio of the average gray value of the target graphic object graphic area to the whole gray range of the infrared image; And calculating and obtaining a comprehensive evalua