Oil Palm Counting and Age Estimation from WorldView-3 Imagery and LiDAR Data Using an Integrated OBIA Height Model and Regression Analysis

Table 2

Age estimation analysis from different height and regression models examined over two kernel sizes.

Block	Number of pixels for each oil palm	Height model			Actual planted age (year)	Regression models
		Min height (meters)	Max height (meters)	Average height (meters)
						Linear (LRF)	Polynomial (PRF)	Exponential (ERF)
A	Single center point (1 × 1 kernel)	0.1	6.17	3.71	4	3.50	3.76	5.21
	3 × 3 kernel	1.9	5.1	3.99		3.81	4.08	4.85
B	Single center point (1 × 1 kernel)	0.05	5.28	2.46	3	2.30	3.41	4.11
	3 × 3 kernel	1.5	3.55	2.65		2.24	3.22	3.80
C	Single center point (1 × 1 kernel)	2.3	9.13	5.95	6	5.1	6.35	7.74
	3 × 3 kernel	3.7	7.01	5.6		5.26	6.18	7.31
D1	Single center point (1 × 1 kernel)	2.4	10.80	8.23	8	7.60	8.85	10.5
	3 × 3 kernel	4.99	8.60	7.45		9. 07	8.13	9.82
D2	Single center point (1 × 1 kernel)	8.62	16.5	11.44	9	9.79	9.21	10.91
	3 × 3 kernel	8.62	14	10.23		9.60	8.9	10.66
E	Single center point (1 × 1 kernel)	4.85	21.55	15.30	22	18.52	19.63	20.47
	3 × 3 kernel	12.52	18.95	16.03		18.68	20.29	22.59

The most accurate estimated age to actual age.