System and method for generating an optimized image with scribble-based annotation of images using a machine learning model

1. A processor-implemented method for generating an optimized image with scribble-based annotation using a machine learning model, comprising:

segmenting, using the machine learning model, a received image from at least one of a cloud, or a user device to obtain a classified image using a plurality of classes, wherein each class is represented with a label, wherein the plurality of classes are obtained based on pre-defined weights of the classified image;

displaying, using a graphical user interface, the classified image which specifies the plurality of classes on the classified image with outlines, wherein the outlines on the classified image are generated by the machine learning model;

enabling a user to mark or scribble on the classified image to annotate the plurality of classes if an area on the classified image is not classified into at least one of the plurality of classes;

assigning, using the machine learning model, a color mask for each scribbled area after receiving the classified image that is marked or scribbled by the user;

computing, using the machine learning model, a loss function for a location of pixels based on the color mask assigned on each scribbled area of the classified image, wherein the loss function identifies whether the classified image matches with the received mark or scribble at the scribbled scribble area on the classified image;

modifying, using the machine learning model, the pre-defined weights for each scribbled area to match the classified image and a determined class on the classified image using the loss function by a learning rate, wherein the learning rate determines a speed of change of the weight;

determining, using the machine learning model, whether the classified image is matched with the determined class on the classified image; and

generating the optimized image if the classified image is matched with the determined class on the classified image.

2. The processor-implemented method of claim 1 , wherein the received image is segmented into the plurality of classes by,

extracting, using an encoder, the received image into at least one feature map to capture a plurality of features, wherein the plurality of features are captured using filters; and

recovering, using a decoder, a resolution of the at least one feature map to obtain the plurality of classes.

3. The processor-implemented method of claim 1 , wherein the outlines that are generated on each class of the classified image are validated using the machine learning model through the graphical user interface.

4. The processor implemented method of claim 1 , wherein the marking or scribbling over the area of the classified image is received from an initial outlined class through the graphical user interface.

5. The processor-implemented method of claim 4 , wherein the marking or scribbling over an area of the plurality of classes is received from the optimized image using the graphical user interface, and the optimized image is again optimized based on the received scribbling and marking over the area of the plurality of classes.

6. The processor-implemented method of claim 1 , wherein the method further comprises,

providing optimized images and corresponding (i) weight of each class and (ii) scribble and mark to the machine learning model as an input data; and

training the machine learning model using the input data to generate a trained machine learning model to generate the optimized image.

7. The processor-implemented method of claim 1 , wherein the method comprises incorporating the received scribbling and marking on the area of the classified image when at least one of (i) each label of the plurality of classes is changed, (ii) a subset of pixels in an area is incorrect, (iii) the class is correct and an assigned label is incorrect, or (iv) all the labels in the classified image are missed.

8. The processor-implemented method of claim 1 , wherein the method comprises modifying the pre-defined weights assigned for each scribbled area based on the loss function using a backpropagation technique.

9. One or more non-transitory computer-readable storage medium storing the one or more sequence of instructions, which when executed by the one or more processors, causes to perform a method of generating an optimized image with scribble-based annotation using a machine learning model, said method comprising:

segmenting, using the machine learning model, a received image from at least one of a cloud, or a user device to obtain a classified image using a plurality of classes, wherein each class is represented with a label, wherein the plurality of classes are obtained based on pre-defined weights of the classified image;

displaying, using a graphical user interface, the classified image which specifies the plurality of classes on the classified image with outlines, wherein the outlines on the classified image are generated by the machine learning model;

enabling a user to mark or scribble on the classified image to annotate the plurality of classes if an area on the classified image is not classified into at least one of the plurality of classes;

assigning, using the machine learning model, a color mask for each scribbled area after receiving the classified image that is marked or scribbled by the user;

computing, using the machine learning model, a loss function for a location of pixels based on the color mask assigned on each scribbled area of the classified image, wherein the loss function identifies whether the classified image matches with the received marking or scribbling at the scribbled area on the classified image;

modifying, using the machine learning model, the pre-defined weights for each scribbled area to match the classified image and a determined class on the classified image using the loss function by a learning rate, wherein the learning rate determines a speed of change of the weight;

determining, using the machine learning model, whether the classified image is matched with the determined class on the classified image; and

generating the optimized image if the classified image is matched with the determined class on the classified image.

10. A system for generating an optimized image with scribble-based annotation using a machine learning model, the system comprising:

a device processor; and

a non-transitory computer-readable storage medium storing one or more sequences of instructions, which when executed by the device processor, causes:

segment, using the machine learning model, a received image from at least one of a cloud, or a user device to obtain a classified image using a plurality of classes, wherein each class is represented with a label, wherein the plurality of classes are obtained based on pre-defined weights of the classified image;

display, using a graphical user interface, the classified image which specifies the plurality of classes on the classified image with outlines, wherein the outlines on the classified image are generated by the machine learning model;

enable a user to mark or scribble on the classified image to annotate the plurality of classes if an area on the classified image is not classified into at least one of the plurality of classes;

assign, using the machine learning model, a color mask for each scribbled area after receiving the classified image that is marked or scribbled by the user;

compute, using the machine learning model, a loss function for a location of pixels based on the color mask assigned on each scribbled area of the classified image, wherein the loss function identifies whether the classified image matches with the received marking or scribbling at the scribbled area on the classified image;

modify, using the machine learning model, the pre-defined weights for each scribbled area to match the classified image and a determined class on the classified image using the loss function by a learning rate, wherein the learning rate determines a speed of change of the weight;

determine, using the machine learning model, whether the classified image is matched with the determined class on the classified image; and

generate the optimized image if the classified image is matched with the determined class on the classified image.

11. The system of claim 10 , wherein the processor is configured to segment the received image into the plurality of classes by,

extracting, using an encoder, the received image into at least one feature map to capture a plurality of features, wherein the plurality of features are captured using filters;

recovering, using a decoder, a resolution of the at least one feature map to obtain the plurality of classes.

12. The system of claim 10 , wherein the processor is configured to validate, using the machine learning model, the outlines that are generated on each class of the classified image through the graphical user interface.

13. The system of claim 10 , wherein the processor is configured to receive the marking or scribbling over the area of the classified image from an initial outlined class through the graphical user interface.

14. The system of claim 13 , wherein the processor is configured to receive the marking or scribbling over the area of the plurality of classes from the optimized image using the graphical user interface and again optimizes based on the received scribbling and marking over the area of the plurality of classes.

15. The system of claim 10 , wherein the processor is configured to

provide optimized images and corresponding (i) weight of each class and (ii) scribble and mark to the machine learning model as an input data;

train the machine learning model using the input data to generate a trained machine learning model to generate the optimized image.

16. The system of claim 10 , wherein the processor is configured to incorporate the received scribbling and marking on the area of the classified image when at least one of (i) each label of the plurality of classes is changed, (ii) a subset of pixels in an area is incorrect, (iii) the class is correct and an assigned label is incorrect, or (iv) all the labels in the classified image are missed.

17. The system of claim 10 , wherein the processor is configured to modify the predefined weights assigned for each scribbled area based on the loss function using a backpropagation technique.