DEPTH-SENSING COMPUTER VISION SYSTEM
In various embodiments, systems and methods for generating a digital representation of a 3D space and objects therein include the steps of providing one or more 3D sensors for generating an output array of pixelwise values, the values including an optical intensity value and a value indicative of an estimated distances to an object within a field of view of the sensor; and processing successive output arrays originating from the 3D sensor(s) into a pixelwise array of depth values, each of the depth values having an associated error metric based at least in part on the associated intensity value.
1 . An image-processing system comprising:
at least one 3D sensor for generating an output array of pixelwise values, the values including an optical intensity value and a value indicative of an estimated distance to an object within a field of view of the sensor;
a processor; and
a depth-compute engine, executable by the processor, for processing successive resulting output arrays originating from the at least one 3D sensor into a pixelwise array of depth values, each of the depth values having an associated error metric based at least in part on the associated intensity value.
2 . The system of claim 1 , wherein the error metric is further based on at least one of sensor noise, dark frame data, ambient light or temperature.
3 . The system of claim 1 , further comprising a controller for operating a machine within a safety envelope, the safety envelope having a volume determined at least in part on the error metric of pixels sensed by the at least one sensor and corresponding to a person proximate to the machine.
4 . The system of claim 1 , further comprising a voltage monitor for over/under monitoring of all voltage rails of the system and interrupting system power upon detection of a failure condition.
5 . A method of generating a digital representation of a 3D space and objects therein, the method comprising the steps of:
providing at least one 3D sensor for generating an output array of pixelwise values, the values including an optical intensity value and a value indicative of an estimated distances to an object within a field of view of the sensor; and
processing successive output arrays originating from the at least one 3D sensor into a pixelwise array of depth values, each of the depth values having an associated error metric based at least in part on the associated intensity value.
6 . The method of claim 5 , wherein the error metric is further based on at least one of sensor noise, dark frame data, ambient light or temperature.
7 . The method of claim 5 , further the step of operating a machine within a safety envelope, the safety envelope having a volume determined at least in part on the error metric of pixels sensed by the at least one sensor and corresponding to a person proximate to the machine.