Xujun Ye

To produce high-quality crops, not only excellent cultivation techniques but also accurate nutrient management techniques are important. In recent years, many nondestructive tools such as the chlorophyll meter “SPAD” and the leaf nitrogen meter “Agri Expert CCN” have been developed for measuring crop leaf chlorophyll and nitrogen contents. However, such devices are still relatively expensive for individual farmers. In this research, we developed a low-cost and small-size camera with built-in LEDs of several specific wavelengths for evaluating the nutrient status of fruit trees. A total of 2 camera prototypes were developed by integrating 3 independent LEDs of specific wavelengths (Camera 1: 950 nm, 660 nm and 560 nm; Camera 2: 950 nm, 660 nm and 727 nm) into the device. In addition, a simple software tool was developed to enable the camera to capture leaf images under different LED lighting conditions. Using the prototypes, we acquired images of apple leaves and investigated the possibility of using the images to estimate the leaf nutrient status indicator SPAD (chlorophyll) and CCN (nitrogen) values obtained using the above-mentioned standard tools. The results indicate that the Camera 1 prototype is superior to the Camera 2 prototype and can potentially be applied to the evaluation of nutrient status in apple leaves.

Reliable information about degree of red coloration in fruit flesh is essential for grading and sorting of red-fleshed apples. We propose a spatially resolved interactance spectroscopy approach as a new rapid and non-destructive technique to estimate degree of red coloration in the flesh of a red-fleshed apple cultivar ‘Kurenainoyume’. A novel measurement system was developed to obtain spatially resolved interactance spectra (190–1070 nm) for apple fruits at eight different light source-detector separation (SDS) distances on fruit surface. Anthocyanins in apple were extracted using a solvent extraction technique, and their contents were quantified with a spectrophotometer. Partial least squares (PLS) regression analyses were performed to develop estimation models for anthocyanin content from spatially resolved interactance spectra. Results showed that the PLS models based on interactance spectra obtained at different SDS distances achieved different predictive accuracy. Further, the system demonstrated the possibility to detect the degree of red coloration in the flesh at specific depths by identifying an optimal SDS distance. This might contribute to provide a detailed profile of the red coloration (anthocyanins) that is unevenly distributed among different depths of the flesh. This new approach may be potentially applied to grading and sorting systems for red-fleshed apples in fruit industry.

Lycopene is one of the most important functional components produced in tomatoes. Traditional methods for determining lycopene content usually rely on destructive sampling followed by chemical analysis. This report introduces the use of a smart device camera combined with a new Android application as a novel, rapid, and nondestructive method for determining tomato lycopene content and fruit grade. We first measured chromaticity values and lycopene content in 60 tomato fruits to develop an effective predictive model for lycopene estimation. Second, a method based on color differences between test fruit and several standard colors was designed as a color classification algorithm for fruit grading. Third, an Android application for lycopene estimation and fruit grading was developed. Finally, application functionality was evaluated using an Android 4.2.2 tablet. The results demonstrated that this new approach was functional and its ability to estimate lycopene content and fruit grade was satisfactory.

In order to prolong the shelf-life of fruits and vegetables, plastic films have been covered on them to improve water retention and keep external bacteria away. It is of great significance to estimate the quality of packaged fruits and vegetables accurately by predicting the shelf-life of them. In this research, hyperspectral technology combined with chemometric methods were employed to estimate the shelf-life of fresh spinach leaves in the same environment. Hyperspectral data covering the range of Vis-NIR (380～1 030 nm) and NIR (874～1 734 nm) were acquired from 300 spinach leaves (75 dishs) which were stored in 4 ℃ among 5 periods (0 d, 2 d, 4 d, 6 d, 8 d). Meanwhile, the chlorophyll contents of all spinach leaves were determined. The mean spectra of 300 spinach leaves (200 leaves in training set and 100 leaves in prediction set) were extracted. And then, principal component analysis (PCA) on the training set of 200 spectra from 5 periods of shelf-life displayed apparent cluster. Partial least-squares discriminant analysis (PLS-DA) models were established according to spectral datas and the virtual levels that we ascribed to the different storage periods previously. The total discriminant accuracy rates of prediction set were 83% (VIS-NIR) and 81% (NIR), respectively. The result indicated that the classification and prediction on the shelf-life of fresh spinach can be realized with hyperspectral technology combined with chemometric methods, which offered a theoretical guidance to evaluate the quality of packaged spinach for consumers, and provided technical supports for the development of instruments used for testing the shelf-life of fruits and vegetables in further study.

This study investigated the capability of hyperspectral imaging technology for monitoring the freshness of chicken meat during storage. Fresh chicken meats were prepared and stored in a refrigerator at 4 °C for 8 days. Hyperspectral images were obtained every 24 hours for meat samples with ImSpector V10E, and the average spectral reflectance data for each sample were extracted. The bacteria in colony forming unit (CFU) for each meat sample was measured by the basic bacterial cultivation techniques. Simple correlation analysis and the two band vegetation index (TBVI) methods were used to develop the spectra-based CFU prediction models. Results indicated that the model with the TBVI based on the wavelengths 700 nm and 650 nm achieved the optimal estimation of CFU on meat surface (R ² =0.9328). The TBVI values based on the above two wavelengths were calculated for all pixels on meat images, and were visualized by displaying the TBVI values for all corresponding pixels on a new image. The predicted CFU for meat samples were then calculated and visualized by incorporating the model into the new TBVI image. The results demonstrated the promising potential of hyperspectral imaging technology for the detection of bacteria on meat surface.