System and method for monitoring stocking shelves

Abstract

A system and method for monitoring stocked shelves is disclosed. A shelf monitor comprising an array of sensors id deployed in proximity to a stocking shelf and so that the sensors face a region on the stocking shelf containing shelved goods. A CPU module is configured to receive outputs from the sensors and execute a detection algorithm to detect a change event in the shelved goods and an algorithm to train the detection algorithm. Shelf monitors can be used in multiple numbers; a master shelf among the shelf monitors serves to aggregate data and local control of the system. A master controller can establish communication with a cloud server, allowing many additional features such as integrating the shelf monitoring system with inventory and point-of-sale databased.

1 . A method of monitoring stock comprising:
 illuminating multiple items in a stocking area with light;   measuring a level of said light at multiple locations in said stocking area;   detecting a change in stock in said stocking area based on said measured level of said light; and   reporting said change.   
     
     
         2 . The method of  claim 1 , wherein said change includes an amount of said stock. 
     
     
         3 . The method of  claim 1 , wherein said change includes an identity of a product in said stock. 
     
     
         4 . The method of  claim 1 , wherein said change includes a misplacement of item of said stock. 
     
     
         5 . The method of  claim 1 , wherein said change includes a filling of the stocking area. 
     
     
         6 . The method of  claim 5 , further comprising measuring and reporting a length of time of said filling. 
     
     
         7 . The method  claim 1 , further comprising:
 recognizing a dimming of said illuminating.   
     
     
         8 . The method of  claim 1 , wherein said detecting is further based on a time length of a change in said measured level of said light. 
     
     
         9 . The method of  claim 1 , further comprising detecting a malfunction of a sensor of said level of said light based on a deviation in said level said light. 
     
     
         10 . The method of  claim 1 , further comprising identifying a location from which a sold object was picked up based on said measuring. 
     
     
         11 . The method of  claim 1 , identify a number of times that a product is picked up before it is sold based on said measuring. 
     
     
         12 . The method of  claim 1 , further comprising calibrating said measuring including measuring an empty state, a full state and a partially full state. 
     
     
         13 . The method of  claim 1 , further wherein said measuring includes a light level under said stock. 
     
     
         14 . The method of  claim 1 , further wherein said measuring includes a light level behind said stock opposite a source of said light. 
     
     
         15 . The method of  claim 1 , wherein said multiple locations includes at least two location covered by a single product of said stock. 
     
     
         16 . The method of  claim 1 , further comprising:
 training a sensor state algorithm to perform said detecting.   
     
     
         17 . The method of  claim 1 , further comprising: wherein said detecting is adjusted based on a time of day. 
     
     
         18 . The method of  claim 1 , wherein said light includes an ambient light. 
     
     
         19 . The method of  claim 1 , wherein said reporting includes activating an indicator in a vicinity of said stocking area. 
     
     
         20 . The method of  claim 1 , wherein said reporting includes sending a message to a human operator.