SpikeSegNet-a deep learning approach utilizing encoder-decoder network with hourglass for spike segmentation and counting in wheat plant from visual imaging

Table 7 Outcome of the approach of spike counting on the test images of 30 plants

Image no	Ground truth	Predicted using model	TP	FP	FN	Precision	Accuracy	F1 score
1	10	9	9	0	0	1.00	1.00	1.00
2	8	7	7	0	0	1.00	1.00	1.00
3	10	9	8	0	0	1.00	1.00	1.00
4	11	10	10	0	1	1.00	0.91	0.95
5	10	10	10	0	0	1.00	1.00	1.00
6	9	9	8	1	0	0.89	0.89	0.94
7	10	9	9	0	0	1.00	1.00	1.00
8	6	6	6	0	0	1.00	1.00	1.00
9	12	11	10	0	1	1.00	0.91	0.95
10	12	12	11	0	1	1.00	0.92	0.96
11	13	12	10	0	1	1.00	0.91	0.95
12	11	10	9	0	1	1.00	0.90	0.95
13	6	6	6	0	0	1.00	1.00	1.00
14	8	8	8	0	0	1.00	1.00	1.00
15	16	15	13	2	1	0.87	0.81	0.90
16	2	2	2	0	0	1.00	1.00	1.00
17	10	10	10	0	0	1.00	1.00	1.00
18	1	1	1	0	0	1.00	1.00	1.00
19	11	10	10	0	0	1.00	1.00	1.00
20	7	7	7	0	0	1.00	1.00	1.00
21	8	7	7	0	1	1.00	0.88	0.93
22	8	7	7	0	1	1.00	0.88	0.93
23	10	10	10	0	0	1.00	1.00	1.00
24	11	10	10	0	1	1.00	0.91	0.95
25	2	2	2	0	0	1.00	1.00	1.00
26	8	8	7	0	0	1.00	1.00	1.00
27	9	8	7	0	1	1.00	0.88	0.93
28	12	10	10	0	2	1.00	0.83	0.91
29	7	7	7	0	0	1.00	1.00	1.00
30	8	8	7	1	0	0.88	0.88	0.93
Average						0.99	0.95	0.97

ISSN: 1746-4811