Target recognition and verification using image processing techniques and/or artifical intelligence

A device, comprising:

• one or more memories;
• and
• one or more processors, operatively coupled to the one or more memories, to:
• receive image data that depicts a tag of a product that is subject to a security check;
• receive product price data that identifies a price of the product;
• identify the tag within the image data that depicts the tag by using one or more object recognition techniques to analyze the image data that depicts the tag;
• identify a first set of tag attributes of the tag;
• determine a risk level associated with the tag based on an analysis of the first set of tag attributes,
• wherein the analysis includes:
• a first analysis to determine whether the first set of tag attributes matches or satisfies a threshold level of similarity with a second set of tag attributes that are stored in association with one or more values that are capable of serving as indicators of the product being priced correctly,
• wherein the one or more values are associated with a price associated with the tag, or
• a second analysis that uses one or more machine learning techniques to analyze the first set of tag attributes to determine a likelihood that the product is priced correctly,
• wherein the one or more processors, when determining the risk level, are to:
• provide, as part of the second analysis, data identifying the first set of tag attributes and the product price data as inputs to a data model to cause the data model to output a set of risk values that represent likelihoods that the product is priced correctly,
• wherein the data model has been trained using the one or more machine learning techniques, and
• determine the risk level based on the set of risk values;
• determine whether the risk level satisfies a threshold risk level;
• select, based on whether the risk level satisfies the threshold risk level, at least one of:
• another device that is to be a recipient of an alert, and
• content that is to be used as part of the alert;
• and
• provide the alert to the other device to cause the other device to perform a first set of actions if the risk level fails to satisfy the threshold risk level or a second set of actions if the risk level satisfies the threshold risk level.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors of a first device, cause the one or more processors to:
• receive, from a reporting device, image data that depicts a tag of a product that is subject to a security check;
• receive product price data that identifies a price of the product;
• identify the tag by using one or more object recognition techniques to analyze the image data that depicts the tag;
• identify a first set of tag attributes of the tag;
• determine a risk level that represents a likelihood of the product being scanned with a correct price by performing an analysis of the first set of tag attributes,
• wherein the analysis includes at least one of:
• a first analysis to determine whether the first set of tag attributes matches or satisfies a threshold level of similarity with a second set of tag attributes that are stored in association with one or more values that are capable of serving as indicators of whether the product was scanned with the correct price,
• wherein the one or more values are associated with a price associated with the tag, or
• a second analysis that uses one or more machine learning techniques to analyze the first set of tag attributes to determine a likelihood of whether the product was scanned with the correct price,
• wherein a defect is present that would prevent the first analysis from being able to determine the risk level, and
• wherein the one or more instructions, that cause the one or more processors to determine the risk level, cause the one or more processors to:
• provide, as part of the second analysis, data identifying the set of tag attributes and the product price data as inputs to a data model to cause the data model to output a set of risk values that represent likelihoods of the product being scanned with the correct price,
• wherein the data model has been trained using the one or more machine learning techniques, and
• determine the risk level based on the set of risk values;
• determine whether the risk level satisfies a threshold risk level;
• and
• provide an alert to a second device to cause the second device to perform a first set of actions based on the risk level not satisfying the threshold risk level or a second set of actions based on the risk level satisfying the threshold risk level.