Table 4 Model parameters and performances for the 2019–2021 combined model

From: Hyperspectral imaging-based classification of rice leaf blast severity over multiple growth stages

Bands	Sample	Trainning	Testing	Model	Average	Micro	Macro
Bands	Sample	Set	Set	Parameters	Accuracy	F1 score	F1 score
All	Level 0	52	43	C = 119.4282	\	\	\
	Level 1	46	114	G = 0.011049	\	\	\
	Level 2	96	49		\	\	\
	Level 3	71	71		\	\	\
	Level 4	65	42		\	\	\
	Total	330	319		88.09%	0.702	0.757
491 nm	Level 0	52	43	C = 315.173	\	\	\
668 nm	Level 1	46	114	G = 16	\	\	\
671 nm^*	Level 2	96	49		\	\	\
	Level 3	71	71		\	\	\
	Level 4	65	42		\	\	\
	Total	330	319		89.47%	0.709	0.759

^* 491, 668 and 671 nm were the combination of selected sensitive wave bands of SRR data acquired in 2019 and 2021

Back to article page

ISSN: 1746-4811

Contact us

Submission enquiries: journalsubmissions@springernature.com