A Vehicle Location Method Based on Road Image Feature Matching

Abstract

The invention discloses a vehicle positioning method based on road surface image feature matching: determining the geographical coordinates of the initial position of the vehicle; taking real-time pictures of the road surface images during the driving process of the vehicle; sequentially performing uniform light processing on the currently photographed two consecutive frames of road surface images; real-time Match the two consecutive frames of road surface images after uniform light processing to obtain the matching point pairs of the two consecutive frames of road surface images; perform vehicle positioning according to the obtained matching point pairs; determine whether the current two frames of images are the last two frames, if so, then end, otherwise repeat the above steps. This method only needs to collect road images in real time while the vehicle is running, and match two consecutive frames of road images to realize autonomous vehicle positioning. The time-consuming and labor-intensive link of full street view, so save time and effort.

Description

technical field

[0001] The invention belongs to the technical field of image processing, and in particular relates to a vehicle positioning method based on road surface image feature matching. Background technique

[0002] Currently, commonly used vehicle positioning technologies include GPS positioning technology and inertial navigation (INS). GPS positioning technology is a passive positioning method, which has defects such as low positioning accuracy, signal blind area, and low output frequency. When the vehicle travels to a no-signal area, GPS cannot meet the requirements; The positioning accuracy cannot meet the requirements. Inertial Navigation (INS) is an autonomous positioning method that obtains the latitude, longitude and acceleration of the vehicle in real time during driving. It has the advantages of all-weather, anti-interference and high instantaneous measurement accuracy. However, its main component, the gyroscope, will drift over time. As a result, its posi...

Examples

Embodiment 1

[0173] In this implementation, a digital CCD camera modeled as SONY XCI-SX100C was installed vertically directly under the vehicle to capture road images in real time, and a 6-meter-long straight road section at the entrance of the School of Information, Chang’an University was selected as a test, shooting 30 frames per second; A total of 279 frames of road images were converted from the video, and the image size is 1280*960.

[0174] The uniform light processing is performed on the currently captured two consecutive frames of road surface images in sequence. The effect of a certain frame image after uniform light processing is as follows: figure 2 shown.

[0175] During the driving process of the vehicle, the captured road surface images are matched in real time, Figure 3-Figure 5 It is the rendering of road surface matching based on SIFT algorithm. When the SIFT algorithm is used, there are 42 groups of images (two consecutive frames of road images) where the number of m...

Embodiment 2

[0184] In this embodiment, a digital CCD camera modeled as SONY XCI-SX100C is vertically installed directly below the vehicle to take real-time images of the road surface, and a section of 12-meter-long turning section at the gate of the School of Information, Chang’an University is selected as a test, and a total of 682 Frame road image, the image size is 1280*960. The positioning process to the vehicle is the same as in embodiment 1, and the positioning track of the final vehicle operation is as follows Figure 9 (b) (the trajectory map of the vehicle running at the pixel level).

[0185] In summary, it can be seen that the method of the present invention realizes autonomous positioning based on road surface image matching, is not easily disturbed by the external environment, and has high accuracy.

technical field

[0001] The invention belongs to the technical field of image processing, and in particular relates to a vehicle positioning method based on road surface image feature matching. Background technique

[0002] Currently, commonly used vehicle positioning technologies include GPS positioning technology and inertial navigation (INS). GPS positioning technology is a passive positioning method, which has defects such as low positioning accuracy, signal blind area, and low output frequency. When the vehicle travels to a no-signal area, GPS cannot meet the requirements; The positioning accuracy cannot meet the requirements. Inertial Navigation (INS) is an autonomous positioning method that obtains the latitude, longitude and acceleration of the vehicle in real time during driving. It has the advantages of all-weather, anti-interference and high instantaneous measurement accuracy. However, its main component, the gyroscope, will drift over time. As a result, its posi...

Examples

Embodiment 1

[0173] In this implementation, a digital CCD camera modeled as SONY XCI-SX100C was installed vertically directly under the vehicle to capture road images in real time, and a 6-meter-long straight road section at the entrance of the School of Information, Chang’an University was selected as a test, shooting 30 frames per second; A total of 279 frames of road images were converted from the video, and the image size is 1280*960.

[0174] The uniform light processing is performed on the currently captured two consecutive frames of road surface images in sequence. The effect of a certain frame image after uniform light processing is as follows: figure 2 shown.

[0175] During the driving process of the vehicle, the captured road surface images are matched in real time, Figure 3-Figure 5 It is the rendering of road surface matching based on SIFT algorithm. When the SIFT algorithm is used, there are 42 groups of images (two consecutive frames of road images) where the number of m...

Embodiment 2

[0184] In this embodiment, a digital CCD camera modeled as SONY XCI-SX100C is vertically installed directly below the vehicle to take real-time images of the road surface, and a section of 12-meter-long turning section at the gate of the School of Information, Chang’an University is selected as a test, and a total of 682 Frame road image, the image size is 1280*960. The positioning process to the vehicle is the same as in embodiment 1, and the positioning track of the final vehicle operation is as follows Figure 9 (b) (the trajectory map of the vehicle running at the pixel level).

[0185] In summary, it can be seen that the method of the present invention realizes autonomous positioning based on road surface image matching, is not easily disturbed by the external environment, and has high accuracy.

Images