Automatic Detection of Hard Exudates in Color Retinal Images Using Dynamic Threshold and SVM Classification: Algorithm Development and Evaluation

Table 2

Overall performance comparison of our proposed algorithm with published algorithms for HE detection on DIARETDB1 database.

Methods	Sensitivity	Specificity	Accuracy
Harangi et al. (2014) [11]	92%	68%	82%
Haloi et al. (2015) [9]	96.54%	93.15%	-
Imani et al. (2016) [39]	89.01%	99.93%	-
Liu et al. (2017) [30]	83%	75%	79%
Rekhi et al. (2017) [31]	91.67%	92.68%	92.13%
Fraz et al. (2017) [10]	92.42%	81.25%	87.72%
Kusakunniran et al. (2018) [40]	89.1%	99.7%	96.2%
Our proposed algorithm	97.5%	97.8%	97.7%