Automated chute fullness detection

1. A method, comprising:

storing (i) a reference image of a chute for receiving objects, and (ii) a region of interest mask corresponding to a location of the chute in a field of view of an image sensor;

at a processor, controlling the image sensor to capture an image of the chute;

applying an illumination adjustment to the image;

selecting, at the processor, a portion of the image according to the region of interest mask;

generating a detection image based on a comparison of the selected portion and the reference image;

determining, based on the detection image, a fullness indicator for the chute; and

providing the fullness indicator to a notification system,

wherein applying the illumination adjustment includes, for each pixel of the selected portion:

determining a difference between the pixel and a corresponding pixel of the reference image;

when the difference exceeds a threshold, determining whether a brightness of the pixel exceeds a brightness threshold; and

when the brightness does not exceed the brightness threshold, adjusting the brightness of the pixel.

2. The method of claim 1 , wherein generating the detection image includes, for each pixel defining the detection image: determining a difference between a first value of a corresponding pixel of the reference image and a second value of a corresponding pixel of the selected portion.

3. The method of claim 2 , wherein determining the fullness indicator includes converting the detection image to a binary image.

4. The method of claim 3 , wherein determining the fullness indicator includes determining a ratio of high-value pixels to low-value pixels in the binary image.

5. The method of claim 3 , wherein determining the fullness indicator includes determining whether the binary image contains any high-value pixels.

6. The method of claim 3 , further comprising:

prior to determining the fullness indicator, detecting regions of high-value pixels; and

discarding a subset of the regions that do not meet a size threshold.

7. The method of claim 1 , further comprising storing an ambient illumination mask;

wherein applying the illumination adjustment includes determining an ambient light level based on a region of the image corresponding to the ambient light mask; and

subtracting the ambient light level from the image.

8. The method of claim 1 , wherein adjusting the brightness includes increasing the brightness of the pixel.

9. A computing device, comprising:

a memory storing (i) a reference image of a chute for receiving objects, and (ii) a region of interest mask corresponding to a location of the chute in a field of view of an image sensor; and

a processor configured to:

control the image sensor to capture an image of the chute;

apply an illumination adjustment to the image;

select a portion of the image according to the region of interest mask;

generate a detection image based on a comparison of the selected portion and the reference image;

determine, based on the detection image, a fullness indicator for the chute; and

providing the fullness indicator to notification system,

wherein the processor is further configured, in order to apply the illumination adjustment, to:

for each pixel of the selected portion:

determine a difference between the pixel and a corresponding pixel of the reference image;

when the difference exceeds a threshold, determine whether a brightness of the pixel exceeds a brightness threshold; and

when the brightness does not exceed the brightness threshold, adjust the brightness of the pixel.

10. The computing device of claim 9 , wherein the processor is configured, in order to generate the detection image, to:

for each pixel defining the detection image:

determine a difference between a first value of a corresponding pixel of the reference image and a second value of a corresponding pixel of the selected portion.

11. The computing device of claim 10 , wherein the processor is configured, in order to determine the fullness indicator, to convert the detection image to a binary image.

12. The computing device of claim 11 , wherein the processor is configured, in order to determine the fullness indicator, to determine a ratio of high-value pixels to low-value pixels in the binary image.

13. The computing device of claim 11 , wherein the processor is configured, in order to determine the fullness indicator, to determine whether the binary image contains any high-value pixels.

14. The computing device of claim 11 , wherein the processor is further configured to:

prior to determining the fullness indicator, detect regions of high-value pixels; and

discard a subset of the regions that do not meet a size threshold.

15. The computing device of claim 9 , wherein the memory stores an ambient illumination mask; and wherein the processor is configured, in order to apply the illumination adjustment, to:

determine an ambient light level based on a region of the image corresponding to the ambient light mask; and

subtract the ambient light level from the image.

16. The computing device of claim 9 , wherein the processor is configured, in order to adjust the brightness, to increase the brightness of the pixel.