Intelligent Detection and Defect Classification of Infusion Bags Based on Support Vector Machine

Fig. 3. Image processing effect of each stage. (a) Original image; (b) first binarization processing image; (c) second binarization processing image; (d) cross median filtering processing image

Download full size

Fig. 4. Discrimination process of one-to-one classification

Download full size

Fig. 5. Partial defect classification detection results. (a) Qualified image; (b) product name offset; (c) product name rotation; (d) product name stain

Download full size

Table 1. Comparison results of three multi-class classification methods
View table
Table 1. Comparison results of three multi-class classification methods
Classificationmethod SVM dividersnumber Recognitionrate /%
OAO 3 96.7
OAA 3 95.8
DAG 2 93.3

Table 2. Test classification results

View table

Table 2. Test classification results

Defect type	Sample	Successnumber	Missingnumber	Wrongnumber	Missingrate /%	Errorrate /%	Positive rate /%
Product name offset	50	49	1	0	2	0	98
Product name rotation	50	49	0	1	0	2	98
Product name stain	50	47	1	2	2	4	94
Total	150	145	2	3	1.3	2	96.7

Table 3. Comparing results of two classification methods
View table
Table 3. Comparing results of two classification methods
Category Proposed method Manual detection
Average operationtime /ms 64 187
Accuracy rate /% 96.7 91

Tools

Get Citation

Copy Citation Text

Dan Li, Yuanyuan Jin, Yan Tong, Guojun Bai, Ming Yang. Intelligent Detection and Defect Classification of Infusion Bags Based on Support Vector Machine[J]. Laser & Optoelectronics Progress, 2019, 56(13): 131502

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites