Xiuqin Rao

Quality and safety assessment of food and agricultural products by hyperspectral fluorescence imaging

Hyperspectral fluorescence imaging (HSFI) is potentially useful for assessing food and agricultural products, because it combines the merits of both hyperspectral imaging and fluorescence spectroscopy. This paper provides an introduction to HSFI: the principle and components of HSFI, calibration and image processing are described. In addition, recent advances in the application of HSFI to food and agricultural product assessment are reviewed, such as contaminant detection, constituent analysis and quality evaluation. Finally, current limitations and likely future development trends are discussed.

Hyperspectral reflectance imaging for detecting citrus canker based on dual-band ratio image classification method

Citrus are one of the major fruit produced in China. Most of this production is exported to Europe for fresh consumption, where consumers increasingly demand best quality. Citrus canker is one of the most devastating diseases that threaten peel of most commercial citrus varieties. The aim of this research was to investigate the potential of using hyperspectral imaging technique for detecting canker lesions on citrus fruit. Navel oranges with cankerous, normal and various common diseased skin conditions including wind scar, thrips scarring, scale insect, dehiscent fruit, phytotoxicity, heterochromatic stripe, and insect damage were studied. The imaging system (400-1000 nm) was established to acquire reflectance images from samples. Region of interest (ROI) spectral feature of various diseased peel areas was analyzed and characteristic wavebands (630, 685, and 720 nm) were extracted. The dual-band reflectance ratio (such as Q720/685) algorithm was performed on the hyperspectral images of navel oranges for differentiating canker from normal fruit skin and other surface diseases. The overall classification success rate was 96.84% regardless of the presence of other confounding diseases. The presented processing approach overcame the presence of stem/navel on navel oranges that typically has been a problematic source for false positives in the detection of defects. Because of the limited sample size, delineation of an optimal detection scheme is beyond the scope of the current study. However, the results showed that two-band ratio (Q685/630) along with the use of a simple threshold value segmentation method for discriminating canker on navel oranges from other peel diseases may be feasible.

Computer vision detection of surface defect on oranges by means of a sliding comparison window local segmentation algorithm

The gray-level image after removing background, (b) the binary image.Display Omitted Successful detections of various types of surface defects.Avoiding additionalimage lightness correction process.Image detection algorithm is novel and practical. Automatic detection of defective oranges by computer vision system is not easy because of the uneven lightness distribution on the surface of oranges. It means that the methods onlydirectly using global segmentation provide unsatisfactory results when orange images present faint defect characters or inhomogeneous surface. The contrast between sound and defective regions can be used to produce more accurate segmentation results, which is more capable of detecting pixels lying around the defect boundary on orange surface based on the local segmentation method. In this paper, we study and propose a sliding comparison window local segmentation algorithm and also presents the detailed image processing procedure including removal of background pixels, image binarization using local segmentation, image subtraction, image morphological modification, removal of stem end pixels for detecting surface defect in an orange gray-level image. This method is an original contribution that allows successful segmentation of various types of surface defects (e.g., insect injury, wind scarring, thrips scarring, scale infestation, canker spot, dehiscent fruit, copper burn, phytotoxicity).The image segmentation algorithm was tested with 1191 samples of oranges. The proposed algorithm was able to correctly detect 97% of the defective orange. Future work will be focused on whole surface and fast on-line inspection.

Spatial frequency domain imaging for detecting bruises of pears

A spatial frequency domain imaging system (SFDI) was developed to estimate the optical properties of biological samples. The system was calibrated by using a series of self-made solid phantoms which covering a wide range of absorption (µa) and reduced scattering coefficients (

Nondestructive determination of optical properties of a pear using spatial frequency domain imaging combined with phase-measuring profilometry

{mu ^{prime}_s}

Relationship between shelf-life and optical properties of Yuanhuang pear in the region of 400–1150 nm

μs′). The relative errors between the reference and calibrated values were regarded as the evaluation parameters for validation effectiveness of the system, the results showed that the maximum relative errors of µa and

Study on Detection Method for Cut Roses Based on Machine Vision

{mu ^{prime}_s}