Features Extraction of Flotation Froth Images and BP Neural Network Soft-Sensor Model of Concentrate Grade Optimized by Shuffled Cuckoo Searching Algorithm

Table 1

Prediction data of soft-sensor model.

Number	Color features			Texture features				Invariant moment (shape features)							Concentrate grade
	Hue	Saturation	Intensity	Energy	Inertia; moment	Entropy	Correlation	𝑛1	𝑛2	𝑛3	𝑛4	𝑛5	𝑛6	𝑛7
1	0.234	0.0235	0.446	0.197	0.114	7.117	0.863	0.047	0.183	0.003	0.605	0.312	0.198	0.359	68.28
2	0.286	0.016	0.514	0.203	0.110	13.633	0.834	0.088	0.162	0.012	0.585	0.384	0.189	0.329	67.3
3	0.310	0.019	0.522	0.195	0.110	11.094	0.853	0.085	0.203	0.038	0.618	0.235	0.164	0.391	68.56
4	0.358	0.023	0.502	0.193	0.107	9.850	0.855	0.047	0.197	0.022	0.613	0.236	0.191	0.397	69.26
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
600	0.229	0.028	0.461	0.192	0.266	7.419	0.934	0.525	0.169	0.078	0.405	0.286	0.073	0.325	69.54