Model performance evaluation for superpixel localization about efficient-compact-fire-detection-cnn

[NeelBhowmik]

Hi,
Apology for the late reply. Did you able to solve the performance issue? Thanks.

[abdullah-al-masud]

No, I couldnt solve the issue. Can you please confirm the way to evaluate superpixel model's performance ?

For example, if I give an image and the SLIC algorithm divides it into 15 parts, among those 15 parts 3 are detected as fire by the model, then how do we evaluate the classification ?

[NeelBhowmik]

Let's consider, total superpixel (SP) = 15

fire (classID = 0) = 3 sp
no-fire (classID = 1) = 12 sp
Ground Truth list of classID; gt = [0,0,0,1,1,1,1,1,1,1,1,1,1,1,1]

Lets consider, model predicts 1st 3 no-fire sp as fire.
model predict list of classID, pred = [0,0,0,0,0,0,1,1,1,1,1,1,1,1,1]

To calculate classification related stats, you need to compare the above two lists (gt, pred).
You can use any standard statistics library, such as sklearn.metrics.classification_report
https://scikit-learn.org/stable/modules/generated/sklearn.metrics.classification_report.html

Thanks