Search

In addition to other surface quality attributes such as size, color and shape, during sorting of harvested apple fruit, early detection of decay is important due to its infectiousness and potential food safety issue. However, automatic and fast inspection of fruit for decay still remains a major problem for the industry. The use of hyperspectral imaging technique makes it possible to perform detection process automatically. Three spectral regions including Vis-NIR (400–1000 nm), Vis (400–780 nm) and NIR (781–1000 nm) were performed using principal component analysis (PCA) to determine the more effective spectral region and PC vector for distinguishing between sound and decayed tissues. Based on the selected PC, loadings corresponding to each wavelength were analyzed to extract key wavelength images in raw hyperspectral data for multispectral image processing. Two sets of multispectral PC score images from Vis-NIR and NIR regions, respectively, were established. To avoid over-segmentation of traditional standard watershed segmentation, global threshold and Ostu, a novel improved watershed segmentation algorithm based on morphological filtering and morphological gradient reconstruction as well as marking constraint were proposed to segment decayed spots on apples. All samples including 220 decayed and 220 sound fruit were used to assess performance of the proposed algorithm. The classification results indicated that 99% of the decayed fruit and 100% of sound fruit were accurately identified by proposed algorithm based on PC3 score images obtained from multispectral PCA of four key wavelengths in NIR region, respectively. This study demonstrated that multispectral images coupled with the improved watershed segmentation algorithm could be a potential approach for detection of early decay on apples. However, further algorithm optimization is still required obtain higher detection accuracy of decayed apples due to zero tolerance for this type of fruit from consumers and processing industries.

Bruise is the most common type of damage to peaches in a major cause of quality loss. However, fast and nondestructive detection of early bruises on peaches is a challenging task. In this study, short-wave near infrared (SW-NIR) and long-wave near infrared (LW-NIR) hyperspectral imaging technologies were observed and compared the ability to discriminate bruised from sound regions. Principal components analysis (PCA) was utilized to select the effective wavelengths for each type of imaging mode. SW-NIR imaging mode was more suitable for detection of early bruises on peaches. A novel improved watershed segmentation algorithm based on morphological gradient reconstruction and marker extraction was developed and applied to the multispectral PC images. The detection results indicated that for all test peaches used in this experiment, 96.5% of the bruised and 97.5% of sound peaches were accurately identified, respectively. A proposed algorithm was superior to the common segmentation methods including Ostu and the global threshold value method. This study demonstrated that SW-NIR hyperspectral imaging coupled with the proposed improved watershed segmentation algorithm could be a potential approach for detection of early bruises on peaches.

The objective of this study was to improve the detection accuracy of soluble solids content (SSC) of apples by integrating spectra and textural features. The spectral data were directly extracted from the region of interest (ROI) of hyperspectral reflectance images of apples over the region of 400–1000 nm, while the textural features were obtained by a texture analysis conducted on the ROI images based on grey-level co-occurrence matrix (GLCM). A new regression method called combined partial least square (CPLS) was proposed to analyze the integrations of spectra and different kinds of textural features. In this algorithm, the score matrix matrices of the spectral data and textural features were obtained by PLS analysis separately and then used together for calibration. The prediction results indicated that the CPLS model developed with the integration of spectra and correlation feature achieved promising results and improved SSC predictions compared with the spectral data when used alone. Next, stability competitive adaptive reweighted sampling (SCARS) was conducted to select informative wavelengths for SSC prediction. The CPLS model based on the integration of SCARS selected spectra and correlation gave better results than those with the full wavelength range. The correlation coefficient and root mean square errors of prediction set and validation set were 0.9327 and 0.641%, 0.913 and 0.6656%, respectively. Hence, the integration of spectra and correlation extracted from hyperspectral reflectance images, coupled with CPLS and SCARS methods, showed a considerable potential for the determination of SSC in apples.

Fruit skin defects may cause fruit spoilage, reduce commodity economic value, and give rise to food quality and safety concerns. Therefore, one of the main tasks of post-harvest processing of fruit is the detection of skin defects by machine vision technology. However, inspection of skin defects on bi-colored fruit varieties by image processing is more difficult because of the high variability of the skin color. This article presents a multispectral detection method for skin defects of bi-colored ‘Pinggu’ peaches based on visible-near infrared (vis–NIR) hyperspectral imaging. Peaches with nine types of skin condition including skin injury, scarring, insect damage, puncture injury, decay, disease spots, dehiscent scarring and anthracnose and normal surface were studied. Principal component analysis (PCA) was used to reduce hyperspectral data dimensionality to select several wavelengths that could potentially be used in an in-line multispectral imaging system. Different defect types produced an obvious feature only in some specific PC images depending on whether the visible light spectrum (425–780 nm), the near infrared spectrum (781–1000 nm), the full-spectrum (400–1000 nm) or only characteristic wavelengths (463, 555, 687, 712, 813, 970 nm or 781, 815, 848 nm) were used. A two-band ratio image ( Q 781/848 ) was successfully used to differentiate defects from a normal surface. Finally, a detection algorithm for skin defects was developed based on a band ratio ( Q 781/848 ) coupled with a simple thresholding method. For the investigated 145 independent test samples with nine skin conditions, an accuracy of 96.6% was obtained, indicating that the proposed multispectral algorithm was effective in differentiating normal and defective bi-colored peaches. The proposed algorithm can be extended to other fruit.

Moldy core is a common disease of apples, but it is difficult to detect because there is no obvious difference in appearance of fruit. In this study, the full-transmittance spectra of apples were collected online with three different orientations at speed of 0.5 m/s using a short-integration-time mode. Spectral measurement orientation has a great influence on the spectral intensity, but no effect on the spectral trends. The spectral intensity of healthy fruit was higher than diseased fruit for all three orientations due to the stronger absorption of damaged tissues. To detect apples with moldy core, four kinds of classification models including naive bayes (NB), linear discriminant analysis (LDA), extreme learning machine (ELM) and support vector machine (SVM) were developed based on the full-transmittance spectra. The results showed that the spectra extracted from medial zone resulted in better detection performance than for intact fruit, and the T2 orientation was more suitable for moldy core detection. The best classification model was built based on the medial zone spectra collected by T2 orientation with the success rate of 90.4 %, 86.9 % and 93.9 % for total, healthy and diseased samples in the validation set. Overall, it is feasible to online detect moldy core with full-transmittance spectroscopy technology, moreover, the spectral acquisition technology of short-integration-time mode can be used to detect internal defect by extracting the effective discrimination information from infected region.

Decay caused by Penicillium spp. fungi is one of the main problems affecting marketing of citrus fruit after harvest because the fungal infection can spread fast from a small number of decayed fruit to the whole consignment. However, the automatic detection of decayed citrus is still a challenge. Early decay of citrus happen on surface peel and present a obvious symptom of water-soaked with cell tissue collapse, which may offer the feasibility of transmittance imaging mode to detect decayed region of citrus. In this study, image processing methods including principal component analysis (PCA), pseudo-color image transformation technology and improved watershed segmentation algorithms (IWSA) were employed to analyze the feasibility of decay detection based on the scanned hyperspectral transmittance images (325−1098 nm) of sound and decayed oranges. The results show that PC3 image is promising for decay segmentation. G components extracted from pseudo-color images of PC3 were selected to enhance image contrasts between decayed and sound tissues, and then decayed regions were segmented perfectly by IWSA whether the defects located on the edge or center position of oranges. However, stem-end tissue had similar features with decayed tissue and therefore were easily misidentified as decayed tissues for those decayed samples with stem-end tissue, and so stem-end identification was carried out. PC2 image and R components extracted from pseudo-color images of PC2 were promising for stem-end identification, then IWSA and morphological parameters were used to extract stem-end region. The stem-end was marked in both operations of decay segmentation and stem-end identification, hence decayed region were further determined for eliminating the misclassification interference of stem-end tissue on decay detection by removing the stem-end region from the operation of decay segmentation. For a validation set including of 84 decayed and 66 sound fruit, the success rates were 93 % and 96 %, respectively, and 94 % for decayed, sound and all fruit. Hyperspectral transmittance imaging offers a novel method for automatic detection of early decayed orange caused by fungus.

Detection of early decay caused by fungal infections in citrus fruit still remains one of the major problems in the post-harvest processing and automatic quality grading. A new combination algorithm by merging multispectral principal component image, bi-dimensional empirical mode decomposition and image reconstruction as well as improved watershed segmentation was proposed to detect the early decay in oranges. Segmented principal component analysis based on three wavelength regions including visible and short wavelength near-infrared (500–1050 nm), visible (500–780 nm) and near-infrared (781–1050 nm) was performed to determine the optimal principal component (PC) image that was used to extract the effective wavelength images by weighting coefficient analysis. Seven wavelength images in the spectral region of 500–1050 nm were finally determined to build the multispectral PC images. The bi-dimensional empirical mode decomposition (BEMD) was used to remove noise in the multispectral PC images and further reconstruct images. An improved watershed segmentation method with morphological gradient reconstruction, marker extraction as well as image amendment, was proposed to segment decay regions in fruit by using the reconstructed multispectral PC images. All samples including 220 each of decayed and sound fruit were utilized to assess classification ability of the proposed combination algorithm. The results indicated that identification accuracies of decayed and sound fruit were 97.3% and 100%, respectively. The multispectral principal component image combining both bi-dimensional empirical mode decomposition and watershed segmentation method can be used as an effective tool for detection of early decayed oranges, and it was also promising for development of a fast and low-cost multispectral imaging system.

Automatic detection of fruit peel defects by a computer vision system is difficult due to the challenges of acquiring images from the surface of spherical fruit and the visual similarity between the stem-ends and the true defects. In this study, oranges with wind scarring, thrips scarring, scale infestation, dehiscent fruit, anthracnose, copper burn, canker spot and normal surface were researched. A lighting transform method based on a low pass Butterworth filter with a cutoff frequency D 0 = 7 was first developed to convert the non-uniform intensity distribution on spherical oranges into a uniform intensity distribution over the whole fruit surface. However, the stem-ends were easily confused with defective areas. In order to solve this problem, different color components (R, G and B) and their combinations were analyzed. It was found that a ratio method and R and G component combination coupled with a big area and elongated region removal algorithm (BER) could be used to differentiate stem-ends from defects effectively. Finally, a processing and classification algorithm based on a simple thresholding method was proposed. The result with 98.9% overall detection rate for the 720 independent sample images indicated that the proposed algorithm was effective in differentiation of normal and defective oranges. The method, however, could not discriminate between different types of defects.