Non-destructive determination of internal defects in chestnut (Castanea mollissima) during postharvest storage using X-ray computed tomography

Wang, Jiahua; Lu, Zelin; Xiao, Xiaofeng; Xu, Mengting; Yuan, Lin; Dai, Honghua; Liu, Xiaodan; Pi, Fuwei; Han, Dong

doi:10.1016/j.postharvbio.2022.112185

Cited by 6 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine vision technology is widely used to detect damage on the surface of agricultural products, but damage to chestnuts often occurs internally, and cannot be detected by machine vision technology when damage occurs inside the chestnut. Good results were obtained using X-ray-CT chestnut defect experiments [3,5]. For example, the accuracy of identifying healthy, severely, and slightly defective chestnuts was 0.929, 0.937, and 0.836, respectively [5].…”

Section: Introductionmentioning

confidence: 99%

“…Good results were obtained using X-ray-CT chestnut defect experiments [3,5]. For example, the accuracy of identifying healthy, severely, and slightly defective chestnuts was 0.929, 0.937, and 0.836, respectively [5]. However, it cannot yet be applied to large-scale chestnut identification and production classification because CT requires theoretical analysis by professionals, which is costly and inefficient.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Feasibility Study of Combining Hyperspectral Imaging with Deep Learning for Chestnut-Quality Detection

Zhong

Tang

et al. 2023

Foods

View full text Add to dashboard Cite

The rapid detection of chestnut quality is a critical aspect of chestnut processing. However, traditional imaging methods pose a challenge for chestnut-quality detection due to the absence of visible epidermis symptoms. This study aims to develop a quick and efficient detection method using hyperspectral imaging (HSI, 935–1720 nm) and deep learning modeling for qualitative and quantitative identification of chestnut quality. Firstly, we used principal component analysis (PCA) to visualize the qualitative analysis of chestnut quality, followed by the application of three pre-processing methods to the spectra. To compare the accuracy of different models for chestnut-quality detection, traditional machine learning models and deep learning models were constructed. Results showed that deep learning models were more accurate, with FD-LSTM achieving the highest accuracy of 99.72%. Moreover, the study identified important wavelengths for chestnut-quality detection at around 1000, 1400 and 1600 nm, to improve the efficiency of the model. The FD-UVE-CNN model achieved the highest accuracy of 97.33% after incorporating the important wavelength identification process. By using the important wavelengths as input for the deep learning network model, recognition time decreased on average by 39 s. After a comprehensive analysis, FD-UVE-CNN was deter-mined to be the most effective model for chestnut-quality detection. This study suggests that deep learning combined with HSI has potential for chestnut-quality detection, and the results are encouraging.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Feasibility Study of Combining Hyperspectral Imaging with Deep Learning for Chestnut-Quality Detection

Zhong

Tang

et al. 2023

Foods

View full text Add to dashboard Cite

show abstract

“…Nondestructive testing technology, particularly industrial computed tomography (CT), holds significance in detecting internal defects and precisely measuring internal structures [6][7][8][9][10][11][12]. Schock et al [13] suggested a combination of high-resolution dual-energy X-ray micro-CT with subsequent advanced image processing steps to derive characteristic parameters for describing real porous systems.…”

Section: Introductionmentioning

confidence: 99%

Numerical investigation of fluid flow behavior in steel cord with lattice Boltzmann methodology: The impacts of microstructure and loading force

Zhao,

Jin,

Fan

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Steel cord materials were found to have internal porous microstructures and complex fluid flow properties. However, current studies have rarely reported the transport behavior of steel cord materials from a microscopic viewpoint. The computed tomography (CT) scanning technology and lattice Boltzmann method (LBM) were used in this study to reconstruct and compare the real three-dimensional (3D) pore structures and fluid flow in the original and tensile (by loading 800 N force) steel cord samples. The pore-scale LBM results showed that fluid velocities increased as displacement differential pressure increased in both the original and tensile steel cord samples, but with two different critical values of 3.3273 Pa and 2.6122 Pa, respectively. The original steel cord sample had higher maximal and average seepage velocities at the 1/2 sections of 3D construction images than the tensile steel cord sample. These phenomena should be attributed to the fact that when the original steel cord sample was stretched, its porosity decreased, pore radius increased, flow channel connectivity improved, and thus flow velocity increased. Moreover, when the internal porosity of tensile steel cord sample was increased by 1 time, lead the maximum velocity to increase by 1.52 times, and the average velocity was increased by 1.66 times. Furthermore, when the density range was determined to be 0–38, the pore phase showed the best consistency with the segmentation area. Depending on the Zou-He Boundary and Regularized Boundary, the relative error of simulated average velocities was only 0.2602 percent.

show abstract

Non-destructive detection and grading of chilling injury-induced lignification of kiwifruit using X-ray computer tomography and machine learning

Wang,

Lin,

et al. 2024

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

Non-destructive determination of internal defects in chestnut (Castanea mollissima) during postharvest storage using X-ray computed tomography

Cited by 6 publications

References 42 publications

Feasibility Study of Combining Hyperspectral Imaging with Deep Learning for Chestnut-Quality Detection

Feasibility Study of Combining Hyperspectral Imaging with Deep Learning for Chestnut-Quality Detection

Numerical investigation of fluid flow behavior in steel cord with lattice Boltzmann methodology: The impacts of microstructure and loading force

Non-destructive detection and grading of chilling injury-induced lignification of kiwifruit using X-ray computer tomography and machine learning

Contact Info

Product

Resources

About