A rapid quantification technique was developed and validated for nondestructively quantifying the soluble solids content of navel orange using NIR spectroscopy in conjunction with Gene expression programming (GEP) algorithm and genetic algorithm (GA).The spectra were recorded in the NIR region using the fiber optic probe method. Four data points of the navel orange NIR spectra were selected by GEP algorithm and genetic algorithm by cross validation. In our experiment, the mean standard error Ribetween the prediction values and the real values in the train group is 0.0256, and the validation group is 0.0194, the results show that the proposed method can be built to a concise easily computed model, which has higher accuracy.

To explore a rapid and non-destructive method for measuring titratable acidity in Gannan Navel Orange, near infrared spectroscopy combined with partial least square method was adopted for building a prediction model. Multiple statistical analysis results show that when the optimal spectral region (950-1330nm) was selected for modeling, the correlation coefficient of the validation set is 0.9085, and RMSECV is 0.0243g/100g. This technique using near infrared spectroscopy is rapid, convenient and nondestructive, and may potentially become a strong tool for the quality evaluation and fruit sorting of navel oranges.

A fast Hough transform for circular object detection is proposed in this paper which can be directly applied to gray level images. This method consists of three major stages. In the first stage, the center positions of circular objects are detected using the gray level Hough transform, which requires no conventional preprocessing such as edge detecting and binarization. The second stage determines the radius of the detected objects by analyzing the radial gradient profile. In order to detect objects with different radius in the same scene, a multi-scale strategy is integrated in the proposed method. Compared with traditional Hough transform, the gray level Hough transform uses a 2-dimensional accumulation map rather than the 3-dimensional one, which results in a dramatic improvement on the computational efficiency. Experiments have been carried out on more than 2000 real-world images and the result shows that 90.3% of the circular objects have been accurately detected, which demonstrate the applicability of the proposed method.