Context-aided machine vision
Various embodiments herein each include at least one of systems, methods, software, and data structures for context-aided machine vision. For example, one method embodiment includes identifying a customer in a shopping area and maintaining an item bin in a computing system of data identifying items the customer has picked up for purchase. This method further includes receiving an image of the customer holding an item and performing item identification processing on the image to identify the item the customer is holding. The item identification processing may be performed based in part on a stored shopping history of the customer indicating items the customer is more likely to purchase. The identified item is then added to the item bin of the customer.
1. A method comprising:
performing item identification processing on a received image to identify an item the customer is holding, the item identification processing based at least in part on a stored shopping history of the customer indicating items the customer is more likely to purchase the item identification processing including applying a multi-dimensional model with data representing a customer likelihood to purchase certain items and to not purchase other items; and
adding an item identified by the item identification processing to an item bin of the customer, the item bin of the customer storing data identifying items the customer has picked up for purchase.
2. The method of claim 1 , wherein the data representing the customer likelihood to purchase certain items and to not purchase other items is, at least in part, a stored shopping history of the customer.
3. The method of claim 1 , wherein the model further includes location, day of the week, and time of day factors indicating a customer likelihood to purchase or not to purchase certain items.
4. The method of claim 1 , wherein the model is utilized by the item identification processing to obtain a weight with regard to each of one or more item possibilities identified by the image processing, wherein the weight with regard to item possibility is a factor for the item identification processing to determine an identity of the item the customer is holding.
5. The method of claim 1 , wherein a model is generated for each of a plurality of customers by processing customer shopping history data stored in a database with at least one machine learning algorithm.
6. The method of claim 5 , wherein the machine learning algorithm is a convolutional neural network algorithm.
7. The method of claim 1 , wherein identifying a customer in a shopping area includes:
capturing an image of at least a portion of a shopping area; and
performing a person recognition function on the image to obtain a customer identity.
8. The method of claim 1 , wherein receiving the image of the customer holding the item includes receiving a cropped image that has been cropped to focus item identification processing on an image portion of a larger image that includes the item.
9. A system comprising:
a computer processor;
a memory device; and
instructions stored on the memory device that are executable by the computer processor to perform data processing activities comprising:
performing item identification processing on a received image to identify an item the customer is holding, the item identification processing based at least in part on a stored shopping history of the customer indicating items the customer is more likely to purchase the item identification processing including applying a multi-dimensional model with data representing a customer likelihood to purchase certain items and to not purchase other items; and
adding an item identified by the item identification processing to an item bin of the customer, the item bin of the customer storing data identifying items the customer has picked up for purchase.
10. The system of claim 9 , wherein the data representing the customer likelihood to purchase certain items and to not purchase other items is, at least in part, a stored shopping history of the customer.
11. The system of claim 9 , wherein the model further includes location, day of the week, and time of day factors indicating a customer likelihood to purchase or not to purchase certain items.
12. The system of claim 9 , wherein the model is utilized by the item identification processing to obtain a weight with regard to each of one or more item possibilities identified by the image processing, wherein the weight with regard to item possibility is a factor for the item identification processing to determine an identity of the item the customer is holding.
13. The system of claim 9 , wherein a model is generated for each of a plurality of customers by processing customer shopping history data stored in a database with at least one machine learning algorithm.
14. The system of claim 13 , wherein the machine learning algorithm is a convolutional neural network algorithm.
15. The system of claim 9 , wherein identifying a customer in a shopping area includes:
capturing an image of at least a portion of a shopping area; and
performing a person recognition function on the image to obtain a customer identity.
16. The system of claim 9 , wherein receiving the image of the customer holding the item includes receiving a cropped image that has been cropped to focus item identification processing on an image portion of a larger image that includes the item.