Method, apparatus, and system for estimating a quality of lane features of a roadway

1. A computer-implemented method for determining an estimated quality of lane features of a roadway comprising:

processing, by a computer vision system, an input image to detect the lane features of the roadway;

determining respective confidence values associated with a plurality of regions of the input image used to detect the lane features, wherein the respective confidence values represent a probability of predicting the lane features from each of the plurality of regions;

performing a classification of the plurality of regions into a plurality of confidence levels based on the respective confidence values;

determining the estimated quality of the lane features based on the classification of the plurality of regions;

specifying a confidence threshold value,

wherein the plurality of confidence levels includes a high confidence level for the plurality of regions associated with the respective confidence values equal to or greater than the confidence threshold value, and a low confidence level for the plurality of regions associated with the respective confidence values less than the confidence threshold value; and

computing a percentage of the plurality of regions that are classified into the high confidence level versus the low confidence value,

wherein the estimated quality of the lane features is based on the percentage.

2. The method of claim 1 , wherein the plurality of regions includes a plurality of pixels, a plurality of cells, or a combination thereof.

3. The method of claim 1 , further comprising:

determining a variation of the estimated quality of the lane features over a distance domain, a temporal domain, or a combination thereof.

4. The method of claim 1 , further comprising:

storing the estimated quality of the lane features in association with a data record of the roadway in a geographic database.

5. The method of claim 1 , further comprising:

determining whether to update a data record of the roadway with the lane features based on the estimated quality of the lane features.

6. The method of claim 1 , further comprising:

configuring a driving operation of an autonomous vehicle based on the estimated quality when the autonomous vehicle approaches or travels on the roadway.

7. The method of claim 1 , further comprising:

storing a history of the estimated quality of the lane features for the roadway; and

detecting a deterioration of paint used to mark the lane features, a condition that obscures the lane features, or a combination thereof based on the history of the estimated quality.

8. The method of claim 1 , further comprising:

determining a routing of a vehicle through the roadway based on the estimated quality.

9. An apparatus for determining an estimated quality of lane features of a roadway comprising:

at least one processor; and

at least one memory including computer program code for one or more programs,

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following,

process, by a computer vision system, an input image to detect the lane features of the roadway;

determine respective confidence values associated with a plurality of regions of the input image used to detect the lane features, wherein the respective confidence values represent a probability of predicting the lane features from each of the plurality of regions;

perform a classification of the plurality of regions into a plurality of confidence levels based on the respective confidence values;

determine the estimated quality of the lane features based on the classification of the plurality of regions;

specify a confidence threshold value, wherein the plurality of confidence levels includes a high confidence level for the plurality of regions associated with the respective confidence values equal to or greater than the confidence threshold value, and a low confidence level for the plurality of regions associated with the respective confidence values less than the confidence threshold value; and

compute a percentage of the plurality of regions that are classified into the high confidence level versus the low confidence value,

wherein the estimated quality of the lane features is based on the percentage.

10. The apparatus of claim 9 , wherein the apparatus is further caused to:

determine a variation of the estimated quality of the lane features over a distance domain, a temporal domain, or a combination thereof.

11. The apparatus of claim 9 , wherein the apparatus is further caused to:

store the estimated quality of the lane features in association with a data record of the roadway in a geographic database.

12. The apparatus of claim 9 , wherein the apparatus is further caused to perform at least one of the following:

determine whether to update a data record of the roadway with the lane features based on the estimated quality of the lane features;

configure a driving operation of an autonomous vehicle based on the estimated quality when the autonomous vehicle approaches or travels on the roadway;

determine a routing of a vehicle through the roadway based on the estimated quality; and

detect a deterioration of paint used to mark the lane features, a condition that obscures the lane features, or a combination thereof based on a history of the estimated quality of the lane features for the roadway.

13. A non-transitory computer-readable storage medium for determining an estimated quality of lane features of a roadway, carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to perform:

processing, by a computer vision system, an input image to detect the lane features of the roadway;

determining respective confidence values associated with a plurality of regions of the input image used to detect the lane features, wherein the respective confidence values represent a probability of predicting the lane features from each of the plurality of regions;

performing a classification of the plurality of regions into a plurality of confidence levels based on the respective confidence values;

determining the estimated quality of the lane features based on the classification of the plurality of regions;

specifying a confidence threshold value, wherein the plurality of confidence levels includes a high confidence level for the plurality of regions associated with the respective confidence values equal to or greater than the confidence threshold value, and a low confidence level for the plurality of regions associated with the respective confidence values less than the confidence threshold value; and

computing a percentage of the plurality of regions that are classified into the high confidence level versus the low confidence value,

wherein the estimated quality of the lane features is based on the percentage.

14. The non-transitory computer-readable storage medium of claim 13 , wherein the apparatus is further caused to perform:

determining a variation of the estimated quality of the lane features over a distance domain, a temporal domain, or a combination thereof.

15. The non-transitory computer-readable storage medium of claim 13 , wherein the apparatus is further caused to perform:

storing the estimated quality of the lane features in association with a data record of the roadway in a geographic database.

16. The non-transitory computer-readable storage medium of claim 13 , wherein the apparatus is further caused to perform at least one of the following:

determining whether to update a data record of the roadway with the lane features based on the estimated quality of the lane features;

configuring a driving operation of an autonomous vehicle based on the estimated quality when the autonomous vehicle approaches or travels on the roadway;

determining a routing of a vehicle through the roadway based on the estimated quality; and

detecting a deterioration of paint used to mark the lane features, a condition that obscures the lane features, or a combination thereof based on a history of the estimated quality of the lane features for the roadway.