Identifying overfilled containers

Patent metadata

Field	Value
Publication number	US-12080040-B2
Application number	US-202318189042-A
Country	US
Kind code	B2
Filing date	Mar 23, 2023
Priority date	Apr 20, 2020
Publication date	Sep 3, 2024
Grant date	Sep 3, 2024

Abstract

Among other things, the techniques described herein include a method for receiving a plurality of images of one or more containers while the one or more containers are being emptied, the plurality of images comprising a training set of images and a validation set of images; labeling each image of the plurality of images as including either an overfilled container or a not-overfilled container; processing each image of the plurality of images to reduce bias of a machine learning model; training, and based on the labeling, the machine learning model using the plurality of images; and optimizing the machine learning model by performing learning against the validation set, the optimized machine learning model being used to generate a prediction for a new image of a container, the prediction indicating whether the container in the new image was overfilled prior to the new container being emptied.

First claim

What is claimed is:

1. A method comprising:
pre-processing, by one or more processors, a plurality of images of one or more containers to extract a hue saturation value (HSV) for each of the plurality of images;
dividing, by the one or more processors and based on the HSV for each of the plurality of images, the plurality of images into different sets of images within the plurality of images;
after dividing the plurality of images into different sets of images, labeling, by the one or more processors, each image of the plurality of images as including either a container associated with a particular condition or a container not associated with the particular condition;
training, by the one or more processors and based on the labeling, a machine learning model using the plurality of labeled images; and
generating, by the one or more processors using the trained machine learning model, a prediction for a new image of a container, the prediction indicating whether the container in the new image is associated with the particular condition.

2. The method of claim 1 , wherein the particular condition comprises at least one of an overfilled container condition, a required repair to the container, a container positioning condition, or unusual activity related to servicing the container.

3. The method of claim 1 , wherein the particular condition comprises an obstruction blocking the container.

4. The method of claim 1 , further comprising:
performing, by the one or more processors and subsequent to the processing, object detection on each image of the plurality of images to identify a respective container within the plurality of images; and
cropping, by the one or more processors, each image of the plurality of images to extract a portion of the image with the respective identified container,
wherein the training of the machine learning model is performed using the images after the cropping has been performed.

5. The method of claim 1 , wherein the different sets of images comprise a first set with images obtained during day and a second set with images obtained during night.

6. The method of claim 1 , wherein each of the different sets of images include images with different corresponding time-stamps, each time-stamp indicating a time when a respective image was obtained by a corresponding vehicle.

7. The method of claim 1 , further comprising:
assigning each of the plurality of images to one of a plurality of classes, wherein:
each class of the plurality of classes is associated with a respective set of the different sets; and
the labeling further comprises labeling the plurality of images according to the plurality of classes.

8. The method of claim 1 , further comprising:
pre-processing, by the one or more processors, the plurality of images to determine geographic locations where the plurality of images were obtained; and
wherein the labeling further comprises labeling the plurality of images according to the geographic locations.

9. The method of claim 1 , wherein:
while a particular container of the one or more containers is being emptied, a sequence of images among the plurality of images are obtained from one or more vehicles; and
the training is performed using the sequence of images.

10. The method of claim 9 , wherein the sequence of images is a video stream.

11. The method of claim 1 , further comprising:
performing, by the one or more processors, object detection on each image of the plurality of images to identify a respective container within the plurality of images; and
masking, by the one or more processors, other objects in the plurality of images,
wherein the training of the machine learning model is performed using the images after the masking has been performed.

12. The method of claim 1 , wherein the processing further comprises:
reducing, by the one or more processors, at least one of glare from headlights or brightness due to sunlight in each image of the plurality of images.

13. A non-transitory computer program product storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:
pre-processing a plurality of images of one or more containers to extract a hue saturation value (HSV) each of the plurality of images;
dividing, based on the HSV for each of the plurality of images, the plurality of images into different sets of images within the plurality of images;
after dividing the plurality of images into different sets of images, labeling each image of the plurality of images as including either a container associated with a particular condition or a container not associated with the particular condition;
training, based on the labeling, a machine learning model using the plurality of labeled images; and
generating, using the trained machine learning model, a prediction for a new image of a container, the prediction indicating whether the container in the new image is associated with the particular condition.

14. A system comprising:
at least one programmable processor; and
a machine-readable medium storing instructions that, when executed by the at least one processor, cause the at least one programmable processor to perform operations comprising:
pre-processing a plurality of images of one or more containers to a extract hue saturation value (HSV) for each of the plurality of images;
dividing, based on the HSV for each of the plurality of images, the plurality of images into different sets of images within the plurality of images;
after dividing the plurality of images into different sets of images, labeling each image of the plurality of images as including either a container associated with a particular condition or a container not associated with the particular condition;
training, based on the labeling, a machine learning model using the plurality of labeled images; and
generating, using the trained machine learning model, a prediction for a new image of a container, the prediction indicating whether the container in the image is associated with the particular condition.

15. A method comprising:
labeling, by one or more processors, each image of a plurality of images as including either a container associated with a particular condition or a container not associated with the particular condition, wherein the particular condition comprises an obstruction blocking the container;
training, by the one or more processors and based on the labeling, a machine learning model using the plurality of labeled images; and
generating, by the one or more processors using the trained machine learning model, a prediction for a new image of a container, the prediction indicating whether the container in the new image is associated with the particular condition.

16. The method of claim 15 , further comprising:
performing, by the one or more processors and subsequent to the processing, object detection on each image of the plurality of images to identify a respective container within the plurality of images; and
cropping, by the one or more processors, each image of the plurality of images to extract a portion of the image with the respective identified container,
wherein the training of the machine learning model is performed using the images after the cropping has been performed.

17. The method of claim 16 , further comprising:
pre-processing, by the one or more processors, a plurality of images of one or more containers to extract a hue saturation value (HSV) for each of the plurality of images; and
dividing, by the

Assignees

• Heil Co

Inventors

• Swaroop Prem
• Kamble Atish P
• Dev Bodhayan

Classifications

• G06V10/776

  Validation; Performance evaluation · CPC title

• G06V10/7747

  Organisation of the process, e.g. bagging or boosting · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06F18/2148

  characterised by the process organisation or structure, e.g. boosting cascade · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title