Table 2 Comparison of 5 cell-detection algorithms.

From: Fast and accurate automated recognition of the dominant cells from fecal images based on Faster R-CNN

Model

mAP

Dur/image

AP

RBC

WBC

Mid

PYO

R-CNN

0.64

14.9 s

0.80

0.74

0.77

0.26

Fast R-CNN

0.66

4.2 s

0.81

0.77

0.79

0.28

Faster R-CNN

0.80

517 ms

0.89

0.81

0.78

0.72

R-FCN

0.81

468 ms

0.90

0.80

0.80

0.73

SSD300

0.63

78 ms

0.475

0.615

0.621

0.821

SSD512

0.74

123 ms

0.630

0.752

0.654

0.812

YOLOv3

0.58

149 ms

0.548

0.628

0.655

0.486

Cascade R-CNN

0.69

263 ms

0.551

0.629

0.775

0.815

PCA-Faster-R-CNN

0.84

723 ms

0.92

0.85

0.81

0.78

  1. Dataset C was used to validate the average precision.
Back to article page