Author Affiliations
State Key Laboratory of Precision Measuring Technology & Instruments, School of Precision Instruments and Optoelectronics Engineering, Tianjin University, Tianjin 300072, Chinashow less
Fig. 1. Typical periodic texture image and spectrum characteristic. (a) Fabric image; (b) 3D spectrum with a series of characteristic peaks
Fig. 2. Simulation texture image and spectrum characteristic. (a) Simulation texture image; (b) 3D spectrum; (c) transverse spectrum and characteristic peaks
Fig. 3. Simulation results of texture image with uniformity error and spectrum characteristic. (a) Simulation texture image; (b) 3D spectrum; (c) transverse spectrum characteristic peaks and peak diffusion
Fig. 4. Correlativity between inspected uniformity error and preseted uniformity error
Fig. 5. Analysis of the relationship between the texture inclination α and the rotation angle of spectrum characteristic point β. (a) Decomposition diagram of d; (b) spectrum characteristic analysis
Fig. 6. User interface in the static density test software. (a) Texture image to be detected; (b) 2D spectrum and feature marker; (c) rotation-corrected image; (d) feature marker in the corrected image; (e) control of execution and output texture detection results
Fig. 7. Histogram analysis of error of measurement results. (a)Data measured by proposed method; (b) data measured by the method in Ref. [20]
Fig. 8. Real-time acquisition and inspection system. (a) Structure of acquisition system; (b) display interface of inspection result
Fig. 9. Fabric uniformity inspection results. (a)~(d) Fabric of uniform texture; (e)~(h) fabric stretched along transverse direction
Fig. 10. Paper counts based on spectrum characteristics. (a) Cross-section image of paper; (b) preprocessing image; (c) 3D spectrum; (d) spectrum characteristics
Fig. 11. Wood texture inspection based on spectrum characteristics. (a) Wood texture image; (b) preprocessing image; (c) 3D spectrum; (d) spectrum characteristics
Fig. 12. Inspection of microstructural arrays based on spectrum characteristics. (a)Texture image of microstructural arrays; (b) preprocessing image; (c) 3D spectrum; (d) spectrum characteristics
Methods | Spline | Cubic | Bilinear | Polyfit |
---|
Maximum error /% | 0.27 | 0.32 | 0.50 | 0.51 | Median error /% | 0.10 | 0.15 | 0.20 | 0.20 | Average error /% | 0.15 | 0.17 | 0.25 | 0.26 |
|
Table 1. Statistical results of measurement errors by different optimization methods
No. | P1 | Error /% | P2 | Error /% |
---|
| Manual | Automatic | Manual | Automatic |
---|
1 | 86.1 | 85.9 | 1.3 | 99.6 | 99.2 | 0.6 | 2 | 68.5 | 68.0 | 0.7 | 71.8 | 71.5 | 1.1 | 3 | 79.6 | 80.0 | 0.5 | 87.5 | 88.2 | 0.6 | 4 | 78.6 | 78.5 | 0.8 | 87.1 | 86.5 | 0.1 | 5 | 81.3 | 81.1 | 0.4 | 88.5 | 88.2 | 0.6 | 6 | 86.5 | 87.1 | 0.6 | 77.5 | 77.9 | 0.6 | 7 | 87.4 | 86.9 | 0.5 | 84.3 | 84.0 | 0.4 | 8 | 85.5 | 85.7 | 0.6 | 84.0 | 84.5 | 1.2 | 9 | 67.5 | 67.8 | 0.7 | 63.6 | 64.0 | 0.6 | 10 | 83.3 | 83.5 | 0.4 | 73.9 | 73.4 | 1.2 | 11 | 84.9 | 84.7 | 1.1 | 79.0 | 78.4 | 1.3 | 12 | 84.6 | 85.0 | 0.5 | 75.8 | 76.0 | 0.3 | 13 | 78.6 | 79.0 | 0.5 | 81.6 | 82.1 | 0.5 | 14 | 69.1 | 69.8 | 1.3 | 70.4 | 70.0 | 0.6 | 15 | 75.8 | 76.2 | 0.3 | 82.4 | 82.2 | 0.5 |
|
Table 2. Density measurement results by proposed method
No. | P1 | Error /% | P2 | Error /% |
---|
| Manual | Automatic | Manual | Automatic |
---|
1 | 35.0 | 34.6 | 1.14 | 46.5 | 46.1 | 0.86 | 2 | 42.0 | 41.5 | 1.19 | 55.0 | 55.4 | 0.73 | 3 | 82.5 | 83.0 | 0.61 | 93.5 | 92.2 | 1.39 | 4 | 55.0 | 55.4 | 0.73 | 61.5 | 62.3 | 1.30 | 5 | 59.0 | 59.9 | 1.53 | 82.0 | 83.0 | 1.22 | 6 | 58.0 | 57.7 | 0.52 | 93.0 | 92.3 | 0.75 | 7 | 44.5 | 43.8 | 1.57 | 82.5 | 83.0 | 0.61 | 8 | 57.0 | 57.4 | 0.70 | 79.5 | 79.6 | 0.31 | 9 | 59.5 | 60.0 | 0.84 | 97.0 | 96.9 | 0.10 | 10 | 82.5 | 83.0 | 0.61 | 88.0 | 87.7 | 0.34 | 11 | 60.0 | 59.9 | 0.17 | 101.0 | 101.5 | 0.50 | 12 | 59.5 | 59.2 | 0.50 | 96.0 | 95.8 | 0.21 |
|
Table 3. Density measurement results in Ref. [20]