A new method for discrimination of apple varieties by means of infrared spectroscopy (NIRS) was developed. First, the
characteristic spectra of apple were got through principal component analysis (PCA), the analysis suggested that the
cumulative reliabilities of PC (principal component)1 and PC2 was more than 98%. The 2-dimensions plot was drawn
with the scores of the first and the second principal components; it appeared to provide the best clustering of the vaneties
of apple. PCA compressed thousands of spectral data into several variables that described the body of spectra; the
several variables were applied as inputs to a back propagation neural network with one hidden layer. 75 samples with
three varieties were selected randomly, then they were used to build BP-ANN model. This model had been used to predict
the varieties of 15 unknown samples; the recognition rate of 100% was achieved. This model is reliable and practicable.
So this paper could offer a new approach to the fast discrimination of apple varieties methods.
The spectral analysis technique based on the spectral reflected property to identify object make the real time inspection
of crop nutrition and fast diagnoses come true. Compared with the conventional means of crop nutrition fast diagnoses,
the information acquired by spectral analysis technique is faster and save both time and labor, it is the basic technique
adopted in the precision agriculture which needs to do research on variable fertilization and irrigation. Using spectral
analysis technique to process crop nutrition real time inspection and fast diagnoses is always the popular research in
remote sensing in agriculture. In this paper, in order to find a simple, quick and untouched method to check the level of
nitrogen in canola, the spectral reflectance and SPAD values of the canola leaves of eight regions were measured by an
ASD Field Spec(R) and SPAD 502 chlorophyll meter. Experiment was made on the leaves taken from live canola, and
the relationship between spectral reflectivity and chlorophyll concentration was analyzed. 32 groups of the chlorophyll
concentration data and the reflected spectra data corresponding to them were acquired, also the correlation between red
edge inflexion point position and chlorophyll concentration was analyzed, and the coefficient of 0.986 was got. The
mathematic model between the first derivative of absorption spectra and chlorophyll concentration was established, and
the coefficient of 0.873 was got. Therefore it indicated more that it is possible to use the hyper-spectroscopy remote
sensing to explore the chlorophyll concentration of canola in ration.
In this paper, in order to find a simple, quick and untouched method to check the level of nitrogen in canola, the spectral reflectance and SPAD values of the canola leaves of fifteen regions were measured by an ASD Field Spec(R) and SPAD 502 chlorophyll meter. The measurements were carried out at experiment field in Zhejiang University during growth season of 2004 to 2005. The authors combine the vegetal course of canola, and use SPAD 502 chlorophyll meter to investigate the distributing rule of chlorophyll concentration of the canola; Measure the third ramous of the canola from top, and analyze the variation rule of the chlorophyll concentration in different vegetal course. The spectral reflectivity property of canola leaves in different growing periods was analyzed, the spectral reflectance of the leaves were gradually getting smaller in the visible region and bigger in the near infrared region before flowering, but with reverse change after flowering. The authors also achieve the chlorophyll concentration and reflectance of 32groups of canola leaves and investigate the quantitative relationships between them by correlation analysis. The results reveal that they have good relationship in the range of 510-640nm and 685-720nm, and the peak of the correlation is at the wavelength of 707nm.