Question

The contingency table below is obtained after an image classification. The overall classification accuracy ($O$) is given as

• A. $O=0.795$
• B. $O=0.850$
• C. $O=0.725$
• D. $O=0.754$

Hint:

From a confusion matrix: Overall accuracy $=$ (sum of diagonal) $\big/$ (grand total). Producer’s and user’s accuracies use column- and row-wise normalizations, respectively.

Answer 1

The Correct Option is A

Overall accuracy $O$ is the ratio of correctly classified samples to the total number of samples: \[ O=\frac{\text{trace of confusion matrix}}{\text{sum of all entries}} =\frac{10+25+35+15+20}{132} =\frac{105}{132}=0.79545\approx 0.795. \]