Object detection machine learning

1. A method for classification of objects in an image, the method comprising:

receiving object classification results of the image from one or more classification engines, the object classification results comprise classification of one or more objects and confidence scores associated with the one or more objects;

aggregating the object classification results from the one or more classification engines to generate a list of confidence scores associated with each of the one or more objects, the list of confidence scores comprises one or more confidence scores from one or more classification engines;

calculating an overall certainty score for each of the one or more objects based at least on the list of confidence scores;

prior to receiving object classification results of the image from one or more classification engines, analyzing the image using a trained neural network;

segmenting, using the trained neural network, the image into a plurality of segments based at least on the analysis of the image;

determining one or more candidate classification engines that are historically accurate in classifying each of the plurality of segments based on one or more attributes of each respective segment; and

sending each of the plurality of segments to a respective classification engine determined to have a historically high accuracy above a predetermined threshold for analysis; and

generating a first orchestrated classification result based the overall certainty score for each of the one or more object.

2. The method of claim 1 , wherein the overall certainty score comprises a highest confidence score in the list of confidence scores.

3. The method of claim 1 , wherein the overall certainty score comprises a number of instances of confidence scores, in the list of confidence scores, having a score above a threshold.

4. The method of claim 1 , wherein the overall certainty score comprises an overall probability of accuracy of the confidence scores in the list of confidence scores.

5. The method of claim 1 , further comprising:

validating the first orchestrated classification result, using a trained neural network; and

generating a second orchestrated classification result based at least in part on the validation of the first orchestrated classification result.

6. The method of claim 5 , wherein the trained neural network is trained to identify one or more candidate classification engines that have a high probability of classifying a segment of an image accurately.

7. The method of claim 6 , wherein validating the first orchestrated classification result comprises:

analyzing the image using the trained neural network;

segmenting the image into a plurality of segments based at least on the analysis of the image;

determining one or more candidate classification engines that are historically accurate in classifying each of the plurality of segments based on one or more attributes of each respective segment; and

validating that the overall certainty score for each of the one or more objects is based on one or more classification engines having historically high accuracy.

8. The method of claim 7 , wherein validating the first orchestrated classification result comprises modifying the first orchestrated classification result to generate the second orchestrated classification result based at least on validating that the overall certainty score for each of the one or more objects is based on one or more classification engines having historically high accuracy.

9. A system for classifying objects of an image, the system comprising:

a memory;

one or more processors coupled to the memory, the one or more processors configured to:

segment the image into a plurality of segments based at least on the analysis of the image;

determine one or more candidate classification engines that are historically accurate in classifying each of the plurality of segments based on one or more attributes of each respective segment; and

validate that the overall certainty score for each of the one or more objects is based on one or more classification engines having historically high accuracy generate a first orchestrated classification result based the overall certainty score for each of the one or more object;

receive object classification results of the image from one or more classification engines, the object classification results comprise classification of one or more objects and confidence scores associated with the one or more objects;

aggregate the object classification results from the one or more classification engines to generate a list of confidence scores associated with each of the one or more objects, the list of confidence scores comprises one or more confidence scores from one or more classification engines;

calculate an overall certainty score for each of the one or more objects based at least on the list of confidence scores; and

generate a first orchestrated classification result based the overall certainty score for each of the one or more object.

10. The system of claim 9 , wherein the one or more processors are further configured to:

prior to receiving object classification results of the image from one or more classification engines, analyze the image using a trained neural network;

segment, by the trained neural network, the image into a plurality of segments based at least on the analysis of the image;

determine one or more candidate classification engines that are historically accurate in classifying each of the plurality of segments based on one or more attributes of each respective segment; and

send each of the plurality of segments to a respective classification engine determined to have a historically high accuracy above a predetermined threshold for analysis.

11. The system of claim 9 , wherein the overall certainty score comprises a highest confidence score in the list of confidence scores.

12. The system of claim 9 , wherein the overall certainty score comprises a number of instances of confidence scores, in the list of confidence scores, having a score above a threshold.

13. The system of claim 9 , wherein the overall certainty score comprises a number of instances of confidence scores, in the list of confidence scores, having a score above a threshold.

14. The system of claim 9 , wherein the overall certainty score comprises an overall probability of accuracy of the confidence scores in the list of confidence scores.

15. The system of claim 9 , wherein the one or more processors are further configured to:

validate the first orchestrated classification result, using a trained neural network; and

generate a second orchestrated classification result based at least in part on the validation of the first orchestrated classification result.

16. The system of claim 15 , wherein the one or more processors are further configured to modify the first orchestrated classification result to generate the second orchestrated classification result based at least on validating that the overall certainty score for each of the one or more objects is based on one or more classification engines having historically high accuracy.