Abstract: A system and method for converting an environmental signal surrounding the checkout terminal to a transmission signal is disclosed. The system verifies the accuracy of the transaction at the checkout terminal by analyzing the transmission signal produced by a signal transmission device, which activates upon a detection of a predetermined environmental signal caused around the vicinity of the checkout terminal. The detection of the predetermined environmental signal is performed by a sensor apparatus placed at the checkout terminal which monitors and senses environmental signals during the transaction of an item being purchased at a store.

Abstract: A system detects a transaction outcome by obtaining video data associated with a transaction area and analyzing the video data to obtain at least one video transaction parameter concerning transactions associated with the transaction area. The transaction area can be a video count of items indicated in the video data as detected by an automated item detection algorithm applied to the video data. The system obtains at least one expected transaction parameter concerning an expected transaction that occurs in the transaction area, such as a scan count of items scanned at a point of sale terminal. The system automatically compares the video transaction parameter(s) to the expected transaction parameter(s) to identify a transaction outcome that may indicate fraudulent activity such as sweethearting in a retail environment.

Abstract: A system detects a transaction outcome by obtaining video data associated with a transaction area and analyzing the video data to obtain at least one video transaction parameter concerning transactions associated with the transaction area. The transaction area can be a video count of items indicated in the video data as detected by an automated item detection algorithm applied to the video data. The system obtains at least one expected transaction parameter concerning an expected transaction that occurs in the transaction area, such as a scan count of items scanned at a point of sale terminal. The system automatically compares the video transaction parameter(s) to the expected transaction parameter(s) to identify a transaction outcome that may indicate fraudulent activity such as sweethearting in a retail environment.

Abstract: System and method for detecting a fraudulent activity at a barcode scanner is disclosed. The method issues an alert when the fraudulent activity is confirmed by comparing the visual signature of the item being transacted over the checkout terminal to the model visual signature. The model visual signature is obtained by averaging the collection of visual signature of the item gathered over a period of time. A human validation via a remote processor is employed to confirm the fraudulent activity verified by a computer.

Abstract: A system and method for replaying a security video at the time of a fraudulent incident at a point-of-sale checkout terminal is presented. The video playback takes place in real-time, when the event occurs, allowing personnel to take appropriate measures, corrective or otherwise, to deal with the event. Examples of such events may include when a weight-based security alert is triggered at a self-checkout, when a cashier or customer has missed scanning an item at the checkout, or other possible events where rapid replay of relevant video is required.

Abstract: A system and method for verifying a non-scan item at a self-checkout point-of-sale terminals provided. A merchandise item that isn't scanned at the self-checkout terminal is identified by a detection system. Further, a user helper device is provided to verify the incident by prompting the shopper for an input. The user helper device is operably connected to the self-checkout point-of-sale terminal and the detection system to determine and verify a fraudulent incident occurring during a transaction activity by the shoppers. Further, a store attendant device is provided for human intervention when necessary.

Abstract: Video frames from a video are compressed into a single image or a single data structure that represents a unique visual flowprint or visual signature for a given activity being modeled from the video frames. The flowprint comprises a computed summary of the original pixel values associated with the video frames within the single image and the flowprint is specific to movements occurring within the video frames that are associated with the given activity. In an embodiment, the flowprint is provided as input to a machine-learning algorithm to allow the algorithm to perform object tracking and monitoring from the flowprint rather than from the video frames of the video, which substantially improves processor load and memory utilization on a device that executes the algorithm, and substantially improved responsiveness of the algorithm.

Abstract: System and method for detecting a fraudulent activity at a barcode scanner is disclosed. The method issues an alert when the fraudulent activity is confirmed by comparing the visual signature of the item being transacted over the checkout terminal to the model visual signature. The model visual signature is obtained by averaging the collection of visual signature of the item gathered over a period of time. A human validation via a remote processor is employed to confirm the fraudulent activity verified by a computer.

Abstract: A system and method for verifying a non-scan item at a self-checkout point-of-sale terminal is provided. A merchandise item that isn't scanned at the self-checkout terminal is identified by a detection system. Further, a user helper device is provided to verify the incident by prompting the shopper for an input. The user helper device is operably connected to the self-checkout point-of-sale terminal and the detection system to determine and verify a fraudulent incident occurring during a transaction activity by the shoppers. Further, a store attendant device is provided for human intervention when necessary.

Abstract: Video frames from a video are compressed into a single image or a single data structure that represents a unique visual flowprint or visual signature for a given activity being modeled from the video frames. The flowprint comprises a computed summary of the original pixel values associated with the video frames within the single image and the flowprint is specific to movements occurring within the video frames that are associated with the given activity. In an embodiment, the flowprint is provided as input to a machine-learning algorithm to allow the algorithm to perform object tracking and monitoring from the flowprint rather than from the video frames of the video, which substantially improves processor load and memory utilization on a device that executes the algorithm, and substantially improved responsiveness of the algorithm.

Abstract: A system and method for replaying a security video at the time of a fraudulent incident at a point-of-sale checkout terminal is presented. The video playback takes place in real-time, when the event occurs, allowing personnel to take appropriate measures, corrective or otherwise, to deal with the event. Examples of such events may include when a weight-based security alert is triggered at a self-checkout, when a cashier or customer has missed scanning an item at the checkout, or other possible events where rapid replay of relevant video is required.

Abstract: An image of a cart/basket is captured based on a triggering event associated with a transaction at a transaction terminal. A determination is made whether the cart/basket is empty or nonempty from the image. When the cart/basket is nonempty a determination is made whether the nonempty cart includes one or more legitimate items or saleable items associated with a store. When the cart/basket is nonempty with one or more saleable items remaining in the cart/basket, and alert is raised to suspend completion of the transaction at the transaction terminal for intervention and audit of the transaction.

Abstract: An image of a candidate produce item is received during a transaction at a transaction terminal. A Feature Vector (FV) for the image is produced. Sales data associated with Produce Look Up (PLU) codes is obtained. Bayesian produce recognition engines are provided the FV and the corresponding sales data. Probabilities returned by the engines are evaluated and a pick list of produce items are produced and/or an entered PLU code provided by an operator of the terminal during the transaction for the candidate produce item is verified or identified as counterfeit.

Abstract: A scan avoidance monitoring system detects instances in which a respective customer in a retail environment fails to properly scan or tally one or more retail items for purchase. To avoid generating false positives, such as flagging of events as scan avoidances when they really are not or likely are not scan avoidances, the scan avoidance system can be configured to require a level of confirmation by one or more other analyzers to flag a particular circumstance as a scan avoidance.

Abstract: Video frames from a video are compressed into a single image or a single data structure that represents a unique visual flowprint or visual signature for a given activity being modeled from the video frames. The flowprint comprises a computed summary of the original pixel values associated with the video frames within the single image and the flowprint is specific to movements occurring within the video frames that are associated with the given activity. In an embodiment, the flowprint is provided as input to a machine-learning algorithm to allow the algorithm to perform object tracking and monitoring from the flowprint rather than from the video frames of the video, which substantially improves processor load and memory utilization on a device that executes the algorithm, and substantially improved responsiveness of the algorithm.

Abstract: Video frames from a video are compressed into a single image or a single data structure that represents a unique visual flowprint or visual signature for a given activity being modeled from the video frames. The flowprint comprises a computed summary of the original pixel values associated with the video frames within the single image and the flowprint is specific to movements occurring within the video frames that are associated with the given activity. In an embodiment, the flowprint is provided as input to a machine-learning algorithm to allow the algorithm to perform object tracking and monitoring from the flowprint rather than from the video frames of the video, which substantially improves processor load and memory utilization on a device that executes the algorithm, and substantially improved responsiveness of the algorithm.

Abstract: A machine-learning algorithm is trained on images with a set of diverse items to produce as output feature vectors in a feature-vector space derived for the set. New item images for new items are passed to the algorithm and new feature vectors are projected into the vector space. A classifier for each new item is trained on the new feature vectors to determine whether the new item is new item or is not that new item. During a transaction, an item code scanned for an item and an item image are obtained. The item image is passed to the algorithm, a feature vector is obtained, a corresponding classifier for the item code is retrieved, the feature vector is passed to the classifier, and a determination is provided as to whether the item image and item code matches a specific item that should be associated with the item code.

Abstract: Embodiments herein include novel ways of alerting store personnel when various activities, events, conditions, etc., occur at the checkout in retail establishments. For example, in accordance with one embodiment, the alerting can take place in substantially real-time, when the event occurs, allowing personnel to take appropriate measures, corrective or otherwise, to deal with the detected event. Examples of such events may include situations such as when a person of interest is detected as shopping at the store, when a cashier has missed scanning an item at the checkout, or to alert store personnel if a non-empty shopping cart has exited the store without payment (a.k.a., a cart push-out). Although the discussion below uses a grocery store as an example retail environment, embodiments herein can be used in any type of retail environment.

Abstract: Video frames from a video are compressed into a single image or a single data structure that represents a unique visual flowprint or visual signature for a given activity being modeled from the video frames. The flowprint comprises a computed summary of the original pixel values associated with the video frames within the single image and the flowprint is specific to movements occurring within the video frames that are associated with the given activity. In an embodiment, the flowprint is provided as input to a machine-learning algorithm to allow the algorithm to perform object tracking and monitoring from the flowprint rather than from the video frames of the video, which substantially improves processor load and memory utilization on a device that executes the algorithm, and substantially improved responsiveness of the algorithm.

Abstract: Upon a triggering event, frames of a video feed are analyzed for activity or motion. One frame or a subset of the frames are selected as a keyframe/keyframes based on one or measures associated with one or more metrics derived from the frames. The keyframe/keyframes optimally reflects an activity or degree of motion across the frames. The keyframe/keyframes are then provided as input to a computer vision or machine learning application to improve the efficiency, accuracy, and/or precision of results produced by or decisions made by the application.