LED Array Position Correction Method Based on Fourier Ptychographic Microscopy

2021 ◽  
Vol 41 (4) ◽  
pp. 0411002
Author(s):  
毛海锋 Mao Haifeng ◽  
赵巨峰 Zhao Jufeng ◽  
崔光茫 Cui Guangmang ◽  
吴小辉 Wu Xiaohui
2011 ◽  
Vol 188 ◽  
pp. 208-211
Author(s):  
Y.C. Jiang ◽  
J.F. Wang ◽  
L.J. Zhang ◽  
B.S. Huang

In order to overcome the low efficiency and precision disadvantages of steel plate position manual correction in the NC(numerical control) Flame Cutting, a high efficient steel plate position correction method is proposed by detecting the edge of steel plate images. Based on the development environment of OpenCV, the video of steel plate is captured based on the DirectShow Technology and the edge linear equation of steel plate is obtained by detecting the steel plate image edge with Canny algorithm and the angle between the equation and the x axis is obtained. The steel plate correction algorithm is designed, and based on the deviation angle of the steel plate, and rotation transformation of related processing images, the steel plate corrections is conducted automatically. The experimental testing of steel plate correction is done and the results show that the method can well done the steel plate correction and the error lower than 0.3 degree and can satisfy the flame cutting process requirement and have a great theory and economic value.


2015 ◽  
Vol 719-720 ◽  
pp. 405-410
Author(s):  
Yuan Ming Ding ◽  
Gang Xiang Guo ◽  
Jin Wei Chen

For the problem of serial robot structure error, method of relative position correction was researched. An adaptive filtering correction method based on robot joint position feedback was proposed. The method built a robot joint adaptive filtering corrector (AFC) to correct robot joint feedback. The real spatial position of robot end-effector can be got through the forward kinematics computation with the corrected joint feedback. Thus, the robot structure error is corrected. Weights matrices of AFC were trained with relative positions between calibration points on the standard calibration module. The proposed method of relative positions correction provides a new way for serial robot structure error correction, which does not directly modify the robot kinematics design parameters, dose not need measuring equipment, and can be applied in field. The simulation results got with LabVIEW robot module show that the corrected trajectory of robot end-effector is more approaching to its true trajectory than before, which means the robot structure error is effectively corrected.Keywords: Correction; Relative position; Serial robot; Kinematics


2021 ◽  
Vol 13 (21) ◽  
pp. 4355
Author(s):  
Changda Liu ◽  
Jiawei Qi ◽  
Jie Li ◽  
Qiuhua Tang ◽  
Wenxue Xu ◽  
...  

Shallow-water depth information is essential for ship navigation and fishery farming. However, the accurate acquisition of shallow-water depth has been a challenge for marine mapping. Combining Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) bathymetry data with multispectral data, satellite-derived bathymetry is a promising solution through which to obtain bathymetric information quickly and accurately. This study proposes a photon refraction correction method considering sea-surface undulations to address errors in the underwater photons obtained by the ICESat-2. First, the instantaneous sea surface and beam emission angle are integrated to determine the sea-surface incidence angle. Next, the distance of photon propagation in water is determined using sea-surface undulation and Snell’s law. Finally, position correction is performed through geometric relationships. The corrected photons were combined with the multispectral data for bathymetric inversion, and a bathymetric map of the Yongle Atoll area was obtained. A bathymetric chart was created using the corrected photons and the multispectral data in the Yongle Atoll. Comparing the results of different refraction correction methods with the data measured shows that the refraction correction method proposed in this paper can effectively correct bathymetry errors: the root mean square error is 1.48 m and the R2 is 0.86.


Sign in / Sign up

Export Citation Format

Share Document