Table A2.

Results from Dataset B of several performance metrics for different object-detection techniques at (a) $FPR$ of ${10}^{-6}$ and (b) ${10}^{-5}$. The total number of unique koalas was 25. The best result of each metric is highlighted by an underline and bold style. The second-best result is written in bold style only. The proposed MOBIVLS algorithm performed better than all of the other techniques against all of the metrics used.

No.	Methods	Recall (%)		F1 Score (%)		Koala Count		Avgkdet (%)
		$\mathit{a}$	$\mathit{b}$	$\mathit{a}$	$\mathit{b}$	$\mathit{a}$	$\mathit{b}$	$\mathit{a}$	$\mathit{b}$
1	AAGD [77]	5.1	13.8	9.3	23.6	5	12	6.6	18.4
2	IAAGD [78]	10.2	27.3	16.8	41.73	7	15	12.3	32.8
3	HB-MLCM [79]	7.2	20.5	12.6	33.1	5	11	9.2	25.9
4	ILCM [80]	12.6	$\mathbf{28.7}$	20.9	$\mathbf{43.5}$	8	14	15.1	$\mathbf{34.9}$
5	MLCM [81]	$\mathbf{16.1}$	28	$\mathbf{25.6}$	43	9	15	$\mathbf{19.6}$	33.9
6	MPCM [82]	7.8	19.5	13.3	31.8	7	14	10.2	25.1
7	TMBM [55]	7.2	20.8	11.9	33.6	5	13	8.8	25.1
8	Faster R-CNN	0.3	2.5	0.3	2.6	1	1	0.3	2.9
	[7,39,40]
9	YOLOv2	0.4	4.1	0.5	4.6	1	2	0.4	4.4
	[7,39,41]
10	Combined 2DCNN	1.1	10.6	1.4	14.2	1	6	1.1	11.4
	[7,39]
11	MOBIVLS [75]	$\mathbf{24.6}$	$\mathbf{56.7}$	$\mathbf{33.9}$	$\mathbf{70.8}$	12	21	$\mathbf{28.4}$	$\mathbf{63.8}$