Time filter

Source Type

Wang X.,Nanjing Forestry University | Zhao M.,Nanjing Forestry University | Ju R.,Nanjing Food and Packaging Machinery Institute | Song Q.,Nanjing Forestry University | And 3 more authors.
Computers and Electronics in Agriculture

Although pork freshness is one of the top concerns to consumers, no systems are currently available to the pork industry that could quantitatively predict its spatial distribution in a rapid and nondestructive way. The main objective of this study was to investigate the feasibility of acousto-optical tunable filter (AOTF) based spectral imagery in the visible/near-infrared region for the non-destructive prediction and visualization of the spoilage-indicating chemicals over the surface of intact fresh pork. We developed an AOTF-based spectral imaging system (wavelength range: 550-1000nm) to visualize pork freshness by mapping the predicted total volatile basic nitrogen (TVB-N) content over the surface. Reflectance hyperspectral images of pork loins in packages (n=43) were acquired from day 3 to day 13 post-mortem, and the corresponding TVB-N references were recorded using conventional chemical procedures. The eligible muscle region of interest (EMROI) on a sample surface was auto-segmented, from which the signature spectrum was extracted. After standard normal variate (SNV) filtering, the signature spectra together with their chemical references were fed into a partial least squares regression (PLSR) to create a prediction model on a consecutive spectral range (575-940nm). An analysis of the regression coefficients identified 9 important predictive wavelengths (575, 600, 615, 705, 765, 825, 885, 915, and 935nm). The prediction model was subsequently refined to use the feature wavelengths only. A leave-one-out (LOO) cross-validation showed that the prediction of the TVB-N contents using the refined model was good and had a root mean square error (RMSECV) of 1.94mg/100g and a coefficient of determination (Rcv2) of 0.89. Finally, the freshness distribution over an entire pork surface was visualized by mapping the pixel-wise TVB-N predictions in pseudo-colors based on the refined model. The spatial prediction was also verified in terms of mean and range. The mean values coincided well with their chemical references (with a R2 of 0.81 and a RMSE of 2.58mg/100g), and the range is within reasonable limits (with 95% pixels within 0-50.0mg/100g). The results indicated that the AOTF-based spectral imagery system could be a promising method to predict pork freshness in an in situ test with unprecedented details of the spatial distribution of freshness.Industrial relevance: An AOTF-based VIS/NIR spectral imagery system has the potential for acceptance sampling in meat production plants or for hygienic supervision in the marketplace to predict the freshness of intact chill-stored pork. © 2013 Elsevier B.V. Source

Zhao M.,Nanjing Forestry University | Ju R.,Nanjing Food and Packaging Machinery Institute | Qi L.,Nanjing Forestry University | Wang X.,Nanjing Forestry University
American Society of Agricultural and Biological Engineers Annual International Meeting 2010, ASABE 2010

A rapid growth of the beverage industry has been witnessed in China during the past two decades. Meanwhile, more attention has been paid to product quality with net content as one of its dimensions. Though quantitative filling technology such as flowmeter has been used in the control of net content in beverage filling, equipments monitoring the actual fill level are also needed. A non-contact capacitive method for this purpose is proposed and a non-contact capacitive sensor is developed based on the rationale that beverages are usually of dielectric constants much higher than their bottles, consequently the capacitance of a filled bottle is determined by its fill level, i.e., the higher the fill level of a bottle is, so would be its capacitance. The non-contact capacitive device developed for fill level inspection consists of a capacitive sensor and a control system. The operational effectiveness of the device was tested on beer in glass bottles and water in plastic bottles, with the results showing a relative broad applicability and desirable accuracies. Source

Zhao M.,Nanjing Forestry University | Wang X.,Nanjing Forestry University | Ju R.,Nanjing Food and Packaging Machinery Institute | Cheng Z.,Nanjing Forestry University
American Society of Agricultural and Biological Engineers Annual International Meeting 2010, ASABE 2010

X-ray Imaging has been adopted for identification of Foreign Objects in food products, especially those in metallic packages that are electromagnetic shielding. As template matching is a viable approach to Foreign Object detection in X-ray images, a method of structuring image templates based on wavelet transforms is introduced in this paper for a specific application of detecting and identifying 3 possible kinds of common Foreign Objects of bones, stones and metal wire in a food product of salted duck gizzard, a local Chinese snack food, in vacuumed packages of Aluminum-Polyethylene (AL-PE). First, X-ray images of packaged food containing foreign objects were acquired. Then the images were enhanced by reducing their noises by wavelet transforms by Wavelets of bior3.7, db1, and syms4, respectively and bior3.7 was finally adopted. Later the enhanced images were segmented to binary images using Bimodal method, an Iterative method, and OTSU method, respectively, with OTSU method having the best accuracy. In the end, the size of image templates was determined and 5 groups of templates are built. To test the performance of these templates, a matching was carried out based on extracted invariant moments of wavelet transforms of the templates and images. Results showed that the templates structured are suitable for the purpose of identification of foreign objects in food. Source

Hong G.,Nanjing Forestry University | Hong G.,Nanjing Institute of Technology | Zhao M.,Nanjing Forestry University | Wang X.,Nanjing Forestry University | And 3 more authors.
Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery

Food foreign bodies harm the human health and hold back the food industry development. To detect food by computer vision and image processing technology, the examination has the advantages of non-destruction, high-speed, accuracy and reliable etc. The meat thickness on-line real-time detection system was first designed and constructed, which collected X-ray image of meat and laser image of meat, including filtering, enhancement and other image preprocessing. The X-ray image system of irregular meat was affected by the gray deviations of uneven thickness of meat. A meat thickness measurement method by laser triangular was proposed to eliminate the gray deviations of the X-ray image of the uneven thickness meat. This study established the mathematical model for the thickness of irregular meat, and a thickness measurement system by symmetrical structure of the laser double-triangulation method. Through calibration method two laser image collected by CCD camera were unified by means of the same coordinates. The meat images were synthesized using the image integration methods based on both gray scale method. The three-dimensional scanning experimental result show that the external shape of meat with corresponding profile, gray variation, and the variational thickness of meat image are close to meat thickness. Source

Zhao M.,Nanjing Forestry University | Cheng Z.,Nanjing Forestry University | Ju R.,Nanjing Food and Packaging Machinery Institute | Wang X.,Nanjing Forestry University
American Society of Agricultural and Biological Engineers Annual International Meeting 2010, ASABE 2010

At present, X-ray imaging systems are employed to detect foreign objects in shielded packaged food products. An innovated method is researched in the paper to classify and to locate foreign objects based on matching by invariant moments. Feature vectors constructed by feature invariant moments extracted from template wavelet coefficient matrix were used for recognition and location of foreign objects in X-ray images. Matching between the normalized invariant moment vectors of target image and template images shows that this matching approach by invariant moments is precise and rapid in recognizing foreign objects in X-ray images of shielding-packaged food products. Source

Discover hidden collaborations