Method for Mixed-Particle Classification Based on Convolutional Neural Network

Fig. 7. Images of mixed particles processed by different methods. (a) Mixed particles; (b) Wiener filtering; (c) binarization and hole filling; (d) watershed segmentation; (e) manually fine segmentation

Download full size

Fig. 8. Particle counting results obtained by different classification methods

Download full size

Fig. 9. Detection accuracy of different classification methods

Download full size

Fig. 10. Cumulative distributions of equivalent diameters of particles obtained by different classification methods

Download full size

Fig. 11. Cumulative distributions of aspect ratios for elongated particles obtained by different classification methods

Download full size

Fig. 12. Cumulative distributions of aspect ratios for spherical particles obtained by different classification methods

Download full size

Fig. 13. Cumulative distributions of aspect ratios for irregular particles obtained by different classification methods

Download full size

Table 1. Feature descriptors of particles

View table

Table 1. Feature descriptors of particles

Parameter	Symbol	Description	Category
Perimeter	P	The distance around the boundary of the region
Area	A	The actual number of pixels in the region
Equivalent diameter	D_eq	D_eq=2 $\sqrt[]{\frac{A}{π}}$	1
Major axis	L	The major axis of the external ellipse
Minor axis	S	The minor axis of the external ellipse
Circularity	C	C= $\frac{P}{2 \sqrt[]{A \cdot π}}$
Aspect ratio	A_R	The aspect ratio of minimum bounding rectangle
Boundary irregularity	B_irr	B_irr=2π $(\frac{\sqrt[]{A}}{P} - \sqrt[]{\frac{L^{2} + S^{2}}{2 LS}})$	2
Uniformity	U	The ratio of the outer rectangle to the outer convex polygon
Angular point	A_P	The number of concave and convex points

Table 2. Structure configurations of conv3_x and conv4_x
View table
Table 2. Structure configurations of conv3_x and conv4_x
Stage Output size /(pixel×pixel) Block structure Block count
conv3_x 28×28 $[\begin{array}{l} 1 \times 1 & 128 \\ 3 \times 3 & 128 \\ 1 \times 1 & 512 \end{array}]$ 3
conv4_x 14×14 $[\begin{array}{l} 1 \times 1 & 256 \\ 3 \times 3 & 256 \\ 1 \times 1 & 1024 \end{array}]$ 20

Table 3. Particle sizes measured by different classification methods

View table

Table 3. Particle sizes measured by different classification methods

Method	Diameter of spherical particles /μm				Diameter of irregular particles /μm			Diameter of elongated particles /μm
Method	${D_{n}}_{10}$	${D_{n}}_{50}$	${D_{n}}_{90}$		${D_{n}}_{10}$	${D_{n}}_{50}$	${D_{n}}_{90}$	${D_{n}}_{10}$		${D_{n}}_{50}$	${D_{n}}_{90}$
Ground truth	111.9			122.5					131.1			98.5	136.4	172.9	74.4	96.0	137.6
CNN	110.0			118.5					126.8			103.3	136.4	171.2	74.3	95.8	135.9
SVM_1	110.0			117.2					129.7			84.6	120.9	160.4	73.3	94.4	138.4
SVM_2	110.5			117.3					131.3			75.7	117.8	159.4	69.7	93.2	139.0
BP	105.3			116.1					137.3			84.2	112.0	157.7	76.0	101.1	147.4

Tools

Get Citation

Copy Citation Text

Yang Cai, Mingxu Su, Xiaoshu Cai. Method for Mixed-Particle Classification Based on Convolutional Neural Network[J]. Acta Optica Sinica, 2019, 39(7): 0712002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites