Neural network-based recognition of trade workers present on industrial sites
An example computing platform comprising is configured to (i) receive, via one or more cameras positioned on a construction site, a plurality of images, (ii) detect, within the plurality of images, a plurality of objects being worn by respective workers on the construction site, (iii) select, from the plurality of images, a set of images depicting a particular worker, and (iv) based on the selected set of images depicting the particular worker, determine a plurality of trade probabilities for the particular worker, each trade probability in the plurality of trade probabilities indicating a likelihood that the particular worker belongs to a particular trade from among a plurality of trades.
1. A computing platform comprising:
a network interface;
at least one processor;
non-transitory computer-readable medium; and
program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor such that the computing platform is configured to:
receive, via one or more cameras positioned on a construction site, a set of images depicting a particular worker;
based on the set of images, determine a plurality of trade probabilities for the particular worker, each trade probability in the plurality of trade probabilities indicating a likelihood that the particular worker belongs to a particular trade from among a plurality of trades;
extract a set of background datasets from the set of images;
determine, based on the set of background datasets, a set of contextual probabilities, wherein each respective contextual probability of the set of contextual probabilities indicates a likelihood that a respective background dataset of the set of background datasets indicates a trade-specific context from among a plurality of trade-specific contexts;
based on the plurality of trade probabilities and the set of contextual probabilities, select a particular trading having a highest probability; and
determine, based on the highest probability, that the particular worker belongs to the particular trade.
2. The computing platform of claim 1 , wherein the program instructions that are executable by the at least one processor such that the computing platform is configured to receive the set of images depicting the particular worker comprise program instructions that are executable by the at least one processor such that the computing platform is configured to receive, via execution of a convolutional neural network, the set of images depicting the particular worker.
3. The computing platform of claim 1 , wherein the program instructions that are executable by the at least one processor such that the computing platform is configured to based on the plurality of trade probabilities and the set of contextual probabilities, select a particular trade having a highest probability comprise program instructions that are executable by the at least one processor such that the computing platform is configured to:
based on the plurality of trade probabilities and the set of contextual probabilities, generate a plurality of refined probabilities, wherein a refined probability, of the plurality of refined probabilities, indicates a modified likelihood that the particular worker belongs to the particular trade of the plurality of trades; and
select, from the plurality of refined probabilities, a highest probability.
4. The computing platform of claim 1 , wherein the program instructions that are executable by the at least one processor such that the computing platform is configured to determine a plurality of trade probabilities for the particular worker comprise program instructions that are executable by the at least one processor such that the computing platform is configured to:
detect, within the set of images, one or more objects being worn by the particular worker.
5. The computing platform of claim 4 , wherein the one or more objects being worn by respective workers on the construction site comprise one or more of (i) a hard hat of a given color, (ii) a vest of a given color, or (iii) a sticker attached to a hard hat or a vest.
6. The computing platform of claim 4 , wherein the set of images depicting the particular worker comprises (i) a first object of the one or more objects being worn by the particular worker, and (ii) a second object of the one or more objects being worn by the particular worker.
7. The computing platform of claim 6 , wherein the program instructions that are executable by the at least one processor such that the computing platform is configured to, based on the set of images depicting the particular worker, determine the plurality of trade probabilities for the particular worker comprise program instructions that are executable by the at least one processor such that the computing platform is configured to determine the plurality of trade probabilities for the particular worker based on the first object and the second object, wherein, when worn together, the first object and the second object are indicative of a particular trade of the plurality of trades.
8. The computing platform of claim 1 , further comprising program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor such that the computing platform is configured to:
based on the set of images, determine a plurality of key-point sets, each of the plurality of key-point sets comprising location information indicating key points identified within a depiction of the particular worker in a particular image; and
based on the plurality of key-point sets, determine a plurality of trade-specific activities that appear to be performed by the particular worker.
9. The computing platform of claim 1 , wherein the program instructions that are executable by the at least one processor such that the computing platform is configured to receive the set of images comprise program instructions that are executable by the at least one processor such that the computing platform is configured to receive at least one unprocessed video stream.
10. A non-transitory computer-readable medium, wherein the non-transitory computer-readable medium is provisioned with program instructions that, when executed by at least one processor, cause a computing platform to:
receive, via one or more cameras positioned on a construction site, a set of images depicting a particular worker;
based on the set of images, determine a plurality of trade probabilities for the particular worker, each trade probability in the plurality of trade probabilities indicating a likelihood that the particular worker belongs to a particular trade from among a plurality of trades;
extract a set of background datasets from the set of images;
determine, based on the set of background datasets, a set of contextual probabilities, wherein each respective contextual probability of the set of contextual probabilities indicates a likelihood that a respective background dataset of the set of background datasets indicates a trade-specific context from among a plurality of trade-specific contexts;
based on the plurality of trade probabilities and the set of contextual probabilities, select a particular trading having a highest probability; and
determine, based on the highest probability, that the particular worker belongs to the particular trade.
11. The non-transitory computer-readable medium of claim 10 , wherein the program instructions that, when executed by at least one processor, cause the computing platform to receive the set of images depicting the particular worker comprise program instructions that, when executed by at least one processor, cause the computing platform to receive, via execution of a convolutional neural network, the set of images depicting the particular worker.
12. The non-transitory computer-readable medium of claim 10 , wherein the program instructions that, when executed by at least one processor, cause the computing platform to, based on the plurality of trade probabilities and the set of contextual probabilities, select a particular trade having a highest probability comprise program instructions that, when executed by at least one processor, cause the computing platform to:
based on the plurality of trade probabilities and the set of contextual probabilities, generate a plurality of refined probabilities, wherein a refined probability, of the plurality of refined probabilities, indicates a modified likelihood that the particular worker belongs to the particular trade of the plurality of trades; and
select, from the plurality of refined probabilities, a highest probability.
13. The non-transitory computer-readable medium of claim 10 , wherein the program instructions that, when executed by at least one processor, cause the computing platform to determine a plurality of trade probabilities for the particular worker comprise program instructions that, when executed by at least one processor, cause the computing platform to:
detect, within the set of images, one or more objects being worn by the particular worker.
14. The non-transitory computer-readable medium of claim 13 , wherein the one or more objects being worn by respective workers on the construction site comprise one or more of (i) a hard hat of a given color, (ii) a vest of a given color, or (iii) a sticker attached to a hard hat or a vest.
15. The non-transitory computer-readable medium of claim 13 , wherein the set of images depicting the particular worker comprises (i) a first object of the one or more objects being worn by the particular worker, and (ii) a second object of the one or more objects being worn by the particular worker.
16. The non-transitory computer-readable medium of claim 15 , wherein the program instructions that, when executed by at least one processor, cause the computing platform to, based on the set of images depicting the particular worker, determine the plurality of trade probabilities for the particular worker comprise program instructions that, when executed by at least one processor, cause the computing platform to determine the plurality of trade probabilities for the particular worker based on the first object and the second object, wherein, when worn together, the first object and the second object are indicative of a particular trade of the plurality of trades.
17. The non-transitory computer-readable medium of claim 10 , wherein the non-transitory computer-readable medium is also provisioned with program instructions that, when executed by at least one processor, cause the computing platform to:
based on the set of images, determine a plurality of key-point sets, each of the plurality of key-point sets comprising location information indicating key points identified within a depiction of the particular worker in a particular image; and
based on the plurality of key-point sets, determine a plurality of trade-specific activities that appear to be performed by the particular worker.
18. The non-transitory computer-readable medium of claim 10 , wherein the program instructions that, when executed by at least one processor, cause the computing platform to receive the set of images comprise program instructions that, when executed by at least one processor, cause the computing platform to receive at least one unprocessed video stream.
19. A method carried out by a computing platform, the method comprising:
receiving, via one or more cameras positioned on a construction site, a set of images depicting a particular worker;
based on the set of images, determining a plurality of trade probabilities for the particular worker, each trade probability in the plurality of trade probabilities indicating a likelihood that the particular worker belongs to a particular trade from among a plurality of trades;
extracting a set of background datasets from the set of images;
determining, based on the set of background datasets, a set of contextual probabilities, wherein each respective contextual probability of the set of contextual probabilities indicates a likelihood that a respective background dataset of the set of background datasets indicates a trade-specific context from among a plurality of trade-specific contexts;
based on the plurality of trade probabilities and the set of contextual probabilities, selecting a particular trading having a highest probability; and
determining, based on the highest probability, that the particular worker belongs to the particular trade.
20. The method of claim 19 , wherein determining a plurality of trade probabilities for the particular worker comprises:
detecting, within the set of images, one or more objects being worn by the particular worker.