Systems, methods and devices for monitoring betting activities

A method for monitoring game activities at a gaming table using a camera positioned on the gaming table or proximate thereto, the method comprising:

• capturing image data corresponding to one or more chips positioned in at least one betting area on a gaming surface of the respective gaming table, the camera positioned such that the camera is able to capture the image data pertaining at least to one or more sides of the one or more chips, the camera adapted to determine one or more depth values corresponding to one or more distances from a reference point to the one or more chips, each of the depth values corresponding to the distance to a corresponding chip, the one or more depth values utilized to establish the at least one betting area maintained as coordinate regions in relation to a three-dimensional model of the gaming table;
• pre-processing the captured image data to determine a region of interest including a chip stack region;
• processing the captured image data to individually identify one or more specific chips of the one or more chips within the chip stack region represented in the captured image data, each specific chip being identified through a chip bounding box established around pixels representing the specific chip by estimating a chip value based on machine-vision interpretable features present on the one or more chips, each chip bounding box established using at least the one or more depth values for depth segmentation of the pixels representing the specific chip in the region of interest by conducting three-dimensional background subtraction based on the three-dimensional model of the gaming table to avoid incorporation of the other chips outside of the coordinate regions of the three-dimensional model relating to a betting area corresponding to the chip stack region;
• identifying the one or more chip values by generating one or more histograms, each histogram corresponding with image data in a corresponding chip bounding box of the one or more specific chips, by processing the one or more histograms to obtain one or more sets of features;
• performing feature recognition on each set of features to compare each set of features against a library of pre-defined reference feature sets to estimate the one or more chip values through identifying for each set of features the pre-defined reference feature set that has the greatest similarity to the set of features;
• and
• generating an output data set indicative of, for each chip of the one or more chips, an estimated type or an estimated value, the output data set consumable to establish betting data for the gaming table, the betting data including betting amounts for the at least one betting area.

A non-transitory computer readable medium storing machine-interpretable instructions, which when executed by a processor, cause the processor to perform a method for monitoring game activities at a gaming table using a camera positioned on the gaming table or proximate thereto, the method comprising:

• capturing image data corresponding to one or more chips positioned in at least one betting area on a gaming surface of the respective gaming table, the camera positioned such that the camera is able to capture the image data pertaining at least to one or more sides of the one or more chips, the camera adapted to determine one or more depth values corresponding to one or more distances from a reference point to the one or more chips, each of the depth values corresponding to the distance to a corresponding chip, the one or more depth values utilized to establish the at least one betting area maintained as coordinate regions in relation to a three-dimensional model of the gaming table;
• pre-processing the captured image data to determine a region of interest including a chip stack region;
• processing the captured image data to individually identify one or more specific chips of the one or more chips within the chip stack region represented in the captured image data, each specific chip being identified through a chip bounding box established around pixels representing the specific chip by estimating a chip value based on machine-vision interpretable features present on the one or more chips, each chip bounding box established using at least the one or more depth values for depth segmentation of the pixels representing the specific chip in the region of interest by conducting three-dimensional background subtraction based on the three-dimensional model of the gaming table to avoid incorporation of the other chips outside of the coordinate regions of the three-dimensional model relating to a betting area corresponding to the chip stack region;
• identifying the one or more chip values by generating one or more histograms, each histogram corresponding with image data in a corresponding chip bounding box of the one or more specific chips by processing the one or more histograms to obtain one or more sets of features;
• performing feature recognition on each set of features to compare each set of features against a library of pre-defined reference feature sets to estimate the one or more chip values through identifying for each set of features pre-defined reference feature set that has the greatest similarity to the set of features;
• and
• generating an output data set indicative of, for each chip of the one or more chips, an estimated type or an estimated value, the output data set used to establish betting data for the gaming table, the betting data including betting amounts for the at least one betting area.