Method of configuring a platform for remote support solution using image analysis

1 . A system of training a deep learning model for support of a device in a remote technical support system, the system comprising:

a processing circuitry configured to:

a) receive a plurality of images of a device to be supported;

b) provide a user interface of annotating each image of the plurality of images,

the user interface being configured to, responsive to input from the user, assign annotation data to each respective image of the plurality of images, the annotation data being data indicative of one or more of:

i) one or more image identification labels, and

ii) one or more image subcomponent label tuples, wherein each subcomponent tuple comprises an identification label of a respective subcomponent of the respective image, and at least one of:

i. a status of the respective subcomponent, and

ii. data indicative of a segmentation of the respective subcomponent in the respective image,

thereby giving rise to a plurality of annotated images, and

the user interface being further configured to enable the user to adjust a respective color of one or more device status indicators in the respective image,

thereby enabling utilization of training images representative of device status indicators in varied lighting environments, and giving rise to synthetic images improving training of the deep learning model for classification of device status indicators; and

c) utilize, at least, the plurality of annotated images, to train a deep learning model to classify and/or segment an image to data indicative of at least one of:

i) one or more image identification labels, and

ii) one or more image subcomponent label tuples;

thereby providing a platform enabling automated creation of a remote support service.

2 . The system of claim 1 , wherein the processing circuitry is further configured to:

a) configure a network-based processing circuitry to utilize the trained deep learning model to classify a received image to data indicative of at least one of:

a. one or more image identification labels, and

b. one or more image subcomponent label tuples.

3 . The system of claim 1 , wherein the processing circuitry is configured to perform optical character recognition (OCR) on at least one image of the plurality of images, and to utilize data derivative of the performing the OCR to train the deep learning model.

4 . The system of claim 1 , wherein the user interface is further configured to:

enable the user to modify the respective image or a subcomponent thereof, thereby giving rise to one or more synthetic images; and

wherein the processing circuitry is further configured to utilize the synthetic images to train the deep learning model.

5 . The system of claim 4 , wherein the user interface is configured to enable the user to perform one or more modifications of the respective image that are selected from a group consisting of:

a) changing focus;

b) changing lighting;

c) changing contrast;

d) changing a background behind the supported device;

e) moving a device cable to a device cable attachment point;

f) rotating a position of a subcomponent; and

g) overlaying or modifying image text.

6 . The system of claim 1 , wherein the processing circuitry is further configured to, subsequent to the assigning, responsive to user input, of annotation data to a first image of the plurality of images:

track a displacement, in a second image, of one or more subcomponents that are annotated in the first image; and

assign respective annotation data to the one or more subcomponents in the second image.

7 . The system of claim 6 , wherein the second image is a frame of a video comprising the first image.

8 . A processor-based method of training a deep learning model for support of a device in a remote technical support system, the method comprising:

a) receiving a plurality of images of a device to be supported;

b) providing a user interface of annotating each image of the plurality of images,

the user interface, responsive to input from the user, assigning annotation data to each respective image of the plurality of images, the annotation data being data indicative of one or more of:

i) one or more image identification labels, and

ii) one or more image subcomponent label tuples, wherein each subcomponent tuple comprises an identification label of a respective subcomponent of the respective image, and at least one of:

i. a status of the respective subcomponent, and

ii. data indicative of a segmentation of the respective subcomponent in the respective image,

thereby giving rise to a plurality of annotated images, and

the user interface further enabling a user to adjust a respective color of one or more device status indicators in the respective image,

thereby enabling utilization of training images representative of device status indicators in varied lighting environments, and giving rise to synthetic images improving training of the deep learning model for classification of device status indicators; and

c) utilizing, at least, the plurality of annotated images, to train a deep learning model to classify an image to data indicative of at least one of:

a. one or more image identification labels, and

b. one or more image subcomponent label tuples;

thereby enabling automated creation of a remote support service.

9 . A computer program product comprising a computer readable non-transitory storage medium containing program instructions, which program instructions when read by a processor, cause the processing circuitry to perform a method training a deep learning model for support of a device in a remote technical support system, the method comprising:

a) receiving a plurality of images of a device to be supported;

b) providing a user interface of annotating each image of the plurality of images,

the user interface, responsive to input from the user, assigning annotation data to each respective image of the plurality of images, the annotation data being data indicative of one or more of:

i) one or more image identification labels, and

ii) one or more image subcomponent label tuples, wherein each subcomponent tuple comprises an identification label of a respective subcomponent of the respective image, and at least one of:

i. a status of the respective subcomponent, and

ii. data indicative of a segmentation of the respective subcomponent in the respective image,

thereby giving rise to a plurality of annotated images, and

the user interface further enabling a user to adjust a respective color of one or more device status indicators in the respective image,

thereby enabling utilization of training images representative of device status indicators in varied lighting environments, and giving rise to synthetic images improving training of the deep learning model for classification of device status indicators; and

c) utilizing, at least, the plurality of annotated images, to train a deep learning model to classify an image to data indicative of at least one of:

a. one or more image identification labels, and

b. one or more image subcomponent label tuples;

thereby enabling automated creation of a remote support service.