Construction object detection
View Patent ↗Aspects of the disclosure relate to identifying construction objects. As an example, an image captured by a camera associated with a vehicle as the vehicle is driven along a roadway may be received. This image may be converted into a first channel corresponding to an average brightness contribution from red, blue and green channels of the image. The image may also be converted into a second channel corresponding to a contribution of a color from the red and the green channels of the image. A template may then be used to identify a region of the image corresponding to a potential construction object from the first channel and the second channel.
1. A method of identifying potential construction objects from images, the method comprising:
receiving, by one or more computing devices, an image captured by a camera associated with a vehicle as the vehicle is driven along a roadway;
converting, by the one or more computing devices, the image into a first channel corresponding to an average brightness contribution from red, blue and green channels of the image;
converting, by the one or more computing devices, the image into a second channel corresponding to a contribution of a color from the red and the green channels of the image; and
using, by the one or more computing devices, a template to identify a region of the image corresponding to a potential construction object from the first channel and the second channel, wherein the template includes a first inner area that is situated within a second middle area that is situated within a third outer area, and using the template further includes processing image data within both the first inner area and third outer area to identify the region such that data within the second middle area exclusive of the first inner area is ignored.
2. The method of claim 1 , further comprising:
using a classifier to determine whether the region includes a stripe; and
identifying the potential construction object as a construction object based on the determination of whether the region includes a stripe.
3. The method of claim 1 , wherein identifying the region further includes determining from the first channel whether image data within the first inner area of the template corresponding to the region meets a brightness threshold value.
4. The method of claim 3 , wherein identifying the region further includes determining from the second channel whether a difference in image data values between the first inner area and the third outer area meets a threshold color value.
5. The method of claim 1 , wherein the contribution of color corresponds to at least one of yellow, orange, and red.
6. A system for identifying potential construction objects from images, the system comprising one or more computing devices configured to:
receive an image captured by a camera associated with a vehicle as the vehicle is driven along a roadway;
convert the image into a first channel corresponding to an average brightness contribution from red, blue and green channels of the image;
convert the image into a second channel corresponding to a contribution of a color from the red and the green channels of the image; and
use a template to identify a region of the image corresponding to a potential construction object from the first channel and the second channel wherein the template includes a first inner area that is situated within a second middle area that is situated within a third outer area, and using the template further includes processing image data within both the first inner area and third outer area to identify the region such that data within the second middle area exclusive of the first inner area is ignored.
7. The system of claim 6 wherein the one or more computing devices are further configured to:
using a classifier to determine whether the region includes a stripe; and
identify the potential construction object as a construction object based on the determination of whether the region includes a stripe.
8. The system of claim 6 , wherein the one or more computing devices are further configured to identify the region by determining from the first channel whether image data within the first inner area of the template corresponding to the region meets a brightness threshold value.
9. The system of claim 8 wherein the one or more computing devices are further configured to identify the region by determining from the second channel whether a difference in image data values between image data from the first inner area and the third outer area meets a threshold color value.
10. The system of claim 6 , further comprising the vehicle.
11. A non-transitory, tangible machine readable medium on which instructions are stored, the instructions, when executed by one or more processors, cause the one or more processors to perform a method of identifying potential construction objects from images, the method comprising:
receiving an image captured by a camera associated with a vehicle as the vehicle is driven along a roadway;
converting the image into a first channel corresponding to an average brightness contribution from red, blue and green channels of the image;
converting the image into a second channel corresponding to a contribution of a color from the red and the green channels of the image; and
using a template to identify a region of the image corresponding to a potential construction object from the first channel and the second channel wherein the template includes a first inner area that is situated within a second middle area that is situated within a third outer area, and using the template further includes processing image data within both the first inner area and third outer area to identify the region such that data within the second middle area exclusive of the first inner area is ignored.
12. The medium of claim 11 , wherein identifying the region further includes determining from the first channel whether the first inner area of the template corresponding to the region meets a brightness threshold value.
13. The medium of claim 12 , wherein identifying the region further includes determining from the second channel whether a difference in image data values between the first inner area and the third outer area meets a threshold color value.
14. The medium of claim 11 , further wherein the method further comprises:
using a classifier to determine whether the region includes a stripe; and
identifying the potential construction object as a construction object based on the determination of whether the region includes a stripe.
15. The method of claim 1 , wherein the first inner area, second middle area, and third outer area are rectangular.
16. The method of claim 1 , further comprising using a second template including an inner area, a middle area, and an outer area are rectangular to identify the region, wherein the second template and first template are different sizes, and a ratio of sizes of the inner area of the second template, the middle area of the second template, and the outer area of the second template is a ratio of sizes of the first inner area, second middle area and third middle area.
17. The method of claim 1 , wherein identifying the region further includes determining from the second channel whether a difference in image data values between the first inner area and the third outer area meets a threshold color value.
18. The method of claim 1 , further comprising, after identifying the region, controlling the vehicle based on the potential construction object.
19. The system of claim 6 , wherein identifying the region further includes determining from the second channel whether a difference in image data values between the first inner area and the third outer area meets a threshold color value.
20. The system of claim 6 , wherein the one or more processors are further configured to, after identifying the region, control the vehicle based on the potential construction object.