Testing device and testing method for detecting stitching defect of panoramic camera
A testing device and a testing method for detecting a stitching defect of a panoramic camera are provided. The testing method includes: accessing the panoramic camera to obtain a stitched image, wherein the stitched image includes a chart image corresponding to a chart, wherein the chart includes multiple black stripes and multiple white stripes; generating a defect image marked with the stitching defect according to the chart image; and outputting the defect image.
1. A testing device for detecting a stitching defect of a panoramic camera, comprising:
a chart, comprising a plurality of black stripes and a plurality of white stripes, and configured to be captured by the panoramic camera;
a transceiver, communicatively connected to the panoramic camera; and
a processor, coupled to the transceiver, wherein the processor is configured to:
access the panoramic camera through the transceiver to obtain a stitched image corresponding to the captured chart, wherein the stitched image comprises a chart image corresponding to the chart;
generate a defect image marked with the stitching defect according to the chart image;
output the defect image through the transceiver;
obtain a defect parameter according to the defect image; and
output the defect parameter through the transceiver.
2. The testing device according to claim 1 , wherein the processor is further configured to:
filter the chart image according to a first grayscale value threshold and a second grayscale value threshold to generate the defect image.
3. The testing device according to claim 2 , wherein the first grayscale value threshold is greater than the second grayscale value threshold, wherein the processor is further configured to:
filter out a first noise in the chart image according to the first grayscale value threshold, wherein a first grayscale value of the first noise is greater than the first grayscale value threshold; and
filter out a second noise in the chart image according to the second grayscale value threshold, wherein a second grayscale value of the second noise is less than the second grayscale value threshold.
4. The testing device according to claim 1 , wherein the processor is further configured to:
execute morphology filtering on the chart image to generate the defect image.
5. The testing device according to claim 4 , wherein the morphology filtering comprises:
filtering out a third noise in the chart image according to an erosion operation.
6. The testing device according to claim 4 , wherein the chart image comprises a striped image, wherein the morphology filtering comprises:
filtering out a fourth noise from the chart image, wherein a length of the fourth noise in a first direction is greater than a pixel number threshold, wherein the first direction is perpendicular to the striped image.
7. The testing device according to claim 6 , wherein the processor is further configured to:
receive a pixel resolution corresponding to the panoramic camera through the transceiver; and
determine the pixel number threshold according to the pixel resolution.
8. The testing device according to claim 7 , wherein the pixel resolution is inversely proportional to the pixel number threshold.
9. The testing device according to claim 1 , wherein the processor is further configured to:
execute blob detection on the chart image to detect a blob on the chart image; and
delete the blob from the chart image to generate the defect image.
10. The testing device according to claim 9 , wherein the processor is further configured to:
delete the blob according to an area of the blob.
11. The testing device according to claim 9 , wherein the processor is further configured to:
delete the blob according to a long-to-short axis ratio of the blob.
12. The testing device according to claim 11 , wherein the processor is further configured to:
delete the blob in response to the long-to-short axis ratio being less than a ratio threshold.
13. The testing device according to claim 1 , wherein the defect parameter comprises at least one of:
a defect number, a defect area, and a defect location.
14. The testing device according to claim 1 , wherein the processor obtains the defect parameter according to at least one of:
an image recognition model and a machine learning model.
15. The testing device according to claim 1 , wherein the processor is further configured to:
extract the chart image from the stitched image according to a region of interest.
16. The testing device according to claim 1 , wherein the black stripes comprise a first black stripe and a second black stripe, and the white stripes comprise a first white stripe, wherein the first white stripe is between the first black stripe and the second black stripe, and the first white stripe, the first black stripe, and the second black stripe are parallel to one another.
17. The testing device according to claim 16 , wherein the chart further comprises an alignment stripe, wherein the alignment stripe is not parallel to the first white stripe.
18. The testing device according to claim 1 , further comprising:
a fixing member, fixing the panoramic camera, so that the stitched image comprises the chart image.
19. The testing device according to claim 1 , further comprising:
a light source, projecting light onto the chart.
20. A testing method for detecting a stitching defect of a panoramic camera, comprising:
accessing the panoramic camera to obtain a stitched image corresponding to a chart, wherein the stitched image comprises a chart image corresponding to the chart, wherein the chart comprises a plurality of black stripes and a plurality of white stripes;
generating a defect image marked with the stitching defect according to the chart image;
outputting the defect image;
obtaining a defect parameter according to the defect image; and
outputting the defect parameter.