Research Projects in 2013-2014

Hyperspectral imaging technology for the quality inspection of fish products (SPECTRAFISH) (Postdoc)

Riccioli C, Da-Wen Sun

Sponsors: European Commission, 7th Framework Programme Theme Capacities

In recent years, hyperspectral imaging (HSI) has gained a wide recognition as a non-destructive and fast quality and safety analysis and assessment method for a wide range of food products. The SPECTRAFISH project aims to bridge the lack of a rapid and objective method of noninvasively inspecting based on quality and safety attributes of finfish products by integrating two conventional optical sensing technologies of computer vision and spectroscopy into unique imaging sensors, a hyperspectral imaging system that can provide not only spatial information but also spectral information for each pixel in an image. The development of a hyperspectral imaging device for the automatic, rapid, objective and non-invasive measurement of quality and safety attributes of finfish fillets will be carried out throughout the project.

---

Affordable variable frequency microwave assisted vacuum drying for heat sensitive foods (Postdoc)

Zhang Z, Da-Wen Sun

Sponsors: European Commission, 7th Framework Programme Theme Capacities

Food drying is a main food processing technology and a commond method for food preservation, providing stable ingredients for a variety of foodstuffs. However, traditional drying (e.g. convective hot air drying) would often cause damage of many heat sensitive components like vitamin C, and deteriorate qualities (such as colour, texture and aroma) of many foods like herbs, spices and sea foods. Freeze-drying can provide great-quality dried foods. Freeze drying costs can be 200±500% higher than that of hot air drying in order to achieve the same final moisture content. The present project is to going to develop microwave vacuum drying, as an energy efficient drying method, combining advantages of microwave drying and vacuum drying, by reducing drying time, and drying temperature, to improve the conservation of many heat sensitive food components during drying process. Furthermore, with the aim of solving fixed frequency microwave heating uniformity problems, variable-frequency microwave (VFM) heating will also be explored.

---

MILD-DRY: Novel microwave assisted vacuum drying for heat sensitive foods (Postdoc)

Drummond L, Da-Wen Sun

Sponsors: European Commission, 7th Framework Programme Theme Capacities

Drying is by far the most useful large scale operation method of keeping solid foods safe for long periods of time. However, heat sensitive foods and products that possess excellent quality in terms of taste, aroma, texture, and appearance, pose a major challenge to dry. Cellular tissues containing gas-filled pores tend to collapse when subjected to dehydration, particularly noticeable with prolonged exposure to elevated drying temperatures, such as those used in convective drying. The Mild-Dry project aims to develop a variable frequency microwave vacuum-drying (MWVD) process that will bring together the advantages of these two effective drying methods, combining the speed of microwave drying and the quality preservation of vacuum drying. The system is expected to reduce the drying time of conventional vacuum drying processes resulting in lower energy and running costs, increased product throughput, preservation of the nutritional value and improved quality of heat sensitive dried foods. The project also aims to plan for the commercial scale up of the Mild-Dry system and its subsequent market entry, whereby its uptake will improve the competitiveness of European Small and Medium Enterprises (SMEs) from the dried foods sector.

---

Studies on drying kinetics and energy-conservation strategies of microwave vacuum technology (PhD)

Zhang L, Zhang Z, Drummond L, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Drying characteristics of some fruits and meats under different microwave vacuum drying conditions will be studied systematically. Based on knowledge of moisture and temperature at different drying stages, an optimal drying method will be constructed by adjusting operation parameters (microwave power density, vacuum level) accordingly at every drying stage. The quality of the dried product (moisture content, texture, water activity, rehydration capacity) will be monitored and the use of pre-treatments, such as ultrasound, to maximize heat and mass transfer efficiency, decrease energy usage and improve product quality, will be evaluated.

---

Comparison of three desorption isotherm determination methods on by using microcrystalline cellulose (PhD)

Zhang L, Zhang Z, Drummond L, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme; European Commission, 7th Framework Programme Theme Capacities

Desorption isotherms of Microcrystalline Cellulose (MCC), for water activity ranging from 0.064 to 0.97, were determined by three methods with different dehydration procedures. The gravimetric static method using saturated salt solutions (GSM) was used as a standard method. Moisture content of MCC in GSM was lowered by taking equilibrium with 9 environments of different relative humidity ordered from high to low. In two fast hygrometric instrument methods, MCC were forced to remove moisture by silica gel (SGM) and oven-heating at 60oC (ODM). Time expenditure and repeatability/reproducibility of the three desorption isotherm measurement methods were compared. The experimental data show that these three methods gave quite similar results; however, it is obvious that the two hygrometric instrument methods were capable of recording more data points and consuming quite less time comparing with SSM methods.

---

Study of microwave-vacuum drying of apple slices using improved methods (MEngSc2).

Wang D, Zhang Z, Da-Wen Sun

Sponsors: University College Dublin

Drying of apple slices with improved microwave vacuum drying methods is investigated in the present study. Water puffed out of the foods due to the internal vapor pressure and is kept at the bottom of the vacuum container during microwave vacuum drying (MWVD). In order to achieve the desired moister content ratio, the water is usually removed by evaporation in the normal microwave vacuum drying process. In the present study, improved methods for removing the water are being studied, like using a vacuum line to suck the water at the bottom of the glass container, which at the same time can maintain the desired level of vacuum pressure inside the container. Drying times and qualities of dried apple slices are compared between the improved methods and normal MWVD process.

---

Ultrasound-assisted release of phenolic compounds from oak chips into a model wine (PhD)

Tao Y, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme; European Commission, 7th Framework Programme Theme Capacities

The enhancement of the release of oak-related compounds from oak chips during wine aging with oak chips may interest the winemaking industry. In this study, the 25-kHz ultrasound waves were used to intensify the mass transfer of phenolics from oak chips into a model wine. The influences of acoustic energy density (6.3-25.8 W/L) on the release kinetics of total phenolics were investigated. The results exhibited that the total phenolic yield released was not affected by acoustic energy density significantly during sonication. Furthermore, there was no significant phenolics degradation during sonication in the current experimental range.

---

Evaluation of natural sausage casing modified by combination of surfactant solution and lactic acid (PhD)

Feng CH, Drummond L, Zhang ZH, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

In order to render natural hog casings to withstand the high pressure drop during immersion vacuum cooling (IVC), treatment using two different food additive combinations (modules): Module 1 (soy lecithin (SL) concentration: 1:27.5, soy oil (SO) concentration: 1.25%, lactic acid (L): 19.5 ml/kg NaCl (solid), residence time (T): 75 min) and Module 2 (SL: 1:30, SO: 2.5%, L: 21 ml/kg NaCl, T: 90 min) were applied and sausages were made using modified casing. Burst pressure, tensile strength, maximum rupture force and elongation of cooked treated casing were determined by physical analysis. Histological structures of sausages were observed by light microscopy. Images illustrated that the casings porosity increased after treatments, which facilitated IVC processing of sausages. Compared to 67% of IVC successful rate of sausages made using untreated casing, sausages made using modified casing presented a higher successful rate (no burst) during IVC.

---

Identification of minced lamb meat adulteration using NIR hyperspectral imaging (PhD)

Pu YY, Feng YZ, Kamruzzaman M, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

The potential of NIR hyperspectral imaging for identification and prediction of heart/liver adulteration content (2-56%) in minced lamb meat was evaluated within the spectral range from 950 to 1600 nm. Relative reflectance of mean spectra decreased with increasing adulteration level at 950-1040 and 1200-1300 nm domains. Regression coefficients of PLS was employed for optimal waveband selection. The coefficient of determination for prediction (R2p) was found to be 0.940 and 0.939 with root mean square error (RMSEP) of 3.959% and 3.973% for heat and liver adulteration, respectively, illustrating the feasibility of NIR hyperspectral imaging for rapid and non-destructive detection of meat adulteration.

---

Hyperspectral imaging for inspecting enterobacteriaceae counts in salmon fillets during cold storage (PhD)

He HJ, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme, Irish Research Council for Science, Engineering and Technology

Enterobacteriaceae counts (EBC) is often used as a useful indicator for food safety evaluation. This work was conducted to investigate the potential of near-infrared (NIR) hyperspectral imaging (900−1700 nm) for rapid and non-destructive determination of EBC distribution in farmed salmon fillets during cold storage. Hyperspectral images were acquired at different storage times for all salmon samples and their spectral features were extracted. Partial least squares regression (PLSR) algorithm was used to correlate the spectral data with the reference EBC values measured standard pour plate method. Important wavelengths were then selected based on the correlation coefficients (RC) of PLSR model, leading to seven individual wavelengths of 921, 931, 1145, 1252, 1366, 1628 and 1658 nm. With the seven wavelengths, an optimised model named RC-PLSR was established, resulting in a correlation coefficient of prediction (RP) of 0.940 and root mean square error of prediction (RMSEP) of 0.507. The overall results indicated that NIR hyperspectral imaging technique combined with PLSR analysis could be used for inspecting EBC of salmon fillets during the period of cold storage.

---

Classification of Listeria innocua and Escherichia coli using near-infrared spectroscopy and chemometrics (PhD)

Feng YZ, Downey G, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Classification of Escherichia coli and Listeria innocua using transflectance near infrared (NIR) spectroscopy together with chemometric methods was investigated in this study. NIR spectra were recorded from a series of dilutions of bacterial suspensions in phosphate buffered saline. Calibration models were developed by using partial least squares discriminant analysis (PLS-DA). Besides the full wavelength calibration models, competitive adaptive reweighted sampling (CARS) was employed for the first time to select some important wavelengths so that simplified models could be achieved for classification of different bacterial species. All the samples were correctly classified using only three wavelengths (1884, 1886 and 1890 nm) during classification of E. coli and L. Innocua samples at species level. This study showed that NIRS could be a potential tool for bacteria classification.

---

MILD-DRY: Affordable variable frequency microwave assisted vacuum drying for heat sensitive foods (Postdoc)

Zhang Z, Da-Wen Sun

Sponsors: European Commission, 7th Framework Programme Theme Capacities

Food drying is a main food processing technology and a common method for food preservation, providing stable ingredients for a variety of foodstuffs. Dried foods also hold many advantages, like long storage life, light weight for transportation. However, traditional drying (e.g. convective hot air drying) would often cause damage of many heat sensitive components like vitamin C, and deteriorate qualities (such as colour, texture and aroma) of many foods like herbs, spices and sea foods. Freeze-drying can provide great-quality dried foods. However, it is a cost and time intensive drying method. Freeze drying costs can be 200±500% higher than that of hot air drying in order to achieve the same final moisture content. The present project is to going to develop microwave vacuum drying, as an energy efficient drying method, combining advantages of microwave drying and vacuum drying, by reducing drying time, and drying temperature, to improve the conservation of many heat sensitive food components during drying process. Furthermore, with the aim of solving fixed-frequency microwave heating uniformity problems, variable-frequency microwave (VFM) heating will also be explored. In order to bridge the gap between laboratory research and industrial application, the present project will develop a prototype system for its validation in commercial food drying facilities, and will deliver a model that relates electrical, thermal and physical properties of the selected foods that are dried under vacuum.

---

Prediction of eating qualities of salmon fillet from colour, marbling and surface texture features using hyperspectral imaging technique  (Postdoc)

Wu D, Da-Wen Sun

Sponsors:Irish Research Council for Science, Engineering and Technology

The development of rapid and accurate quality and safety inspection systems with the function of visualizing attribute distribution are important for the Irish salmon industry to ensure the safe production of salmon products during processing operations and the labeling of products correctly related to the quality, safety, authenticity and compliance. With the integration of the main advantages of spectroscopy and imaging, hyperspectral imaging technique can simultaneously acquire spectral and spatial information in one system, leading to its capability of measuring both external physical and morphological characteristics and internal quality attributes from a salmon fillet. In order to do this, a reliable hyperspectral imaging system with quantitative models will be trained, established, and validated for rapid and non-invasive measurement of quality and safety attributes of salmon fillets based on the most critical image features extracted from visible and invisible (near infrared) hyperspectral images of salmon fillets and their reference attribute values, which will be measured by using traditional instruments and sensory analysis.

---

Studies on drying kinetics and energy-conservation strategies of microwave vacuum technology (PhD)

Zhang L, Zhang Z, Drummond L, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Drying characteristics of some fruits and meats under different microwave vacuum drying conditions will be studied systematically. Based on knowledge of moisture and temperature at different drying stages, an optimal drying method will be constructed by adjusting operation parameters (microwave power density, vacuum level) accordingly at every drying stage. The quality of the dried product (moisture content, texture, water activity, rehydration capacity) will be monitored and the use of pre-treatments, such as ultrasound, to maximize heat and mass transfer efficiency, decrease energy usage and improve product quality, will be evaluated.

---

Numerical modeling and experimental analysis of ultrasound assisted freezing of potato spheres (PhD)

Kiani H, Zhang Z, Da-Wen Sun

Sponsors: University of Tehran and the Iranian Ministry of Science

In recent years, innovative methods such as ultrasound assisted freezing have been developed in order to improve the freezing process. In the present study, ultrasound assisted immersion freezing process of potato spheres was evaluated using experimental and numerical approaches. Ultrasound (25 kHz, 890 W m-2) was irradiated (total time of 270 s during the phase change step) for different duty cycles (DCs) (0, 10, 30, 50, 70 and 100%). A finite volume based enthalpy method was used to model the solidification phenomenon and numerical simulation of temperature and liquid fraction profiles was carried out by using OpenFOAM® CFD software. The results showed that ultrasound irradiation could decrease the characteristic freezing time of potatoes. Since ultrasound irradiation increased the heat transfer coefficient but simultaneously generated heat at the surface of the samples, an optimum DC was needed for the shortest freezing time which occurred in the range of 30-70% DC. DCs higher than 70% increased the freezing time.  DCs lower than 30% did not provide significant effects on the freezing time compared to the control sample. The numerical model predicted the characteristic freezing time in accordance with the experimental results. As the numerical simulations provided profiles of temperature and water fraction within potatoes frozen with or without ultrasound, the models can be used to study and control different operation situations, and to improve the understanding of the freezing process.

---

Ultrasound-assisted extraction of phenolic compounds from wine lees (PhD)

Tao Y, Da-Wen Sun

Sponsors: European Commission, 7th Framework Programme Theme Capacities

The influences of different factors, including extraction time, extraction temperature, solvent-to-solid ratio and solvent composition, on the extraction yields of total phenolics and total anthocyanins during ultrasound-assisted extraction from wine lees were investigated in this paper. An ultrasound bath system with the frequency of 40 kHz was used and the acoustic energy density during extraction was identified to 48 W/L. The extraction yields increased markedly in the first 10 min of sonication. In the meantime, the extractions yields increased along with the increase of extraction temperature and solvent-to-solid ratio in the current experimental range. Furthermore, it was found that 50% aqueous ethanol was most effective to extract phenolic compounds from wine lees samples. The experimental range of each factor determined for further optimization of the extraction process included 20-40 min for extraction time, 40-60 oC for extraction temperature, 30:1-50:1 for solvent-to-solid ratio and 40-60% for ethanol concentration.

---

Feasibility of employing vacuum cooling and immersion vacuum cooling to Irish jumbo sausages (PhD)

Feng C, Drummond L, Zhang ZH, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

The effects of vacuum cooling (VC) and immersion vacuum cooling (IVC) on cooling and quality parameters of Irish Jumbo sausages was investigated in the current experiments. Cooling time, mass loss, final achievable sausage core temperature, quality properties (hardness, gumminess and springiness, water holding capacity), core colour, and the evolution of sausage appearance with different storage times were analysed. Results showed that it was difficult for vacuum cooled sausages to reach core temperature below 4 oC. In addition, the cooling time to 10.4 oC core temperature (final achievable temperature) for vacuum cooled sausages was 72.6 min longer than that of IVC samples. In regard to mass loss, product mass loss after VC (7.38%) was significantly higher than that after IVC (1.67%) (P<0.05). Accordingly, moisture content of VC samples (52.20%) was significantly lower than that of IVC products (55.02%) (P<0.05). However, there were no significant differences in relation to instrumental texture, colour and water holding capacity (WHC) (P>0.05). As for appearance changes with time, the images displayed that the initial shrunk and uneven sausage surface after VC and IVC changed to a smoother surface appearance with time. Conclusions can be drawn that it is more suitable to employ immersion vacuum cooling than vacuum cooling to cool these products.

---

Study of drying kinetics of microwave-vacuum drying of apple slices (MEngSc2)

Wang X, Zhang Z, Da-Wen Sun

Sponsors: University College Dublin

Drying is one of the most important preservation processes for apples. Microwave-vacuum drying is an innovative combined drying method, which carries both benefits of microwave drying and vacuum drying. Currently, several vegetables and fruits are successfully microwave- vacuum dried and experimental data fitted using models such as the Lewis, Page, Henderson and Pabis model, etc. This project aims to describe microwave-vacuum drying characteristics of apples and compare the suitability of several models under different operating conditions.

---

Microwave vacuum drying of apple slices (MEngSc2)

Liu R, Zhang Z, Da-Wen Sun

Sponsors: University College Dublin

Colour, texture, vitamin C, sugar, water activity and rehydration ratio of apple slices dried by microwave vacuum drying under different conditions are going to be analysed systematically. Taking the main factors (microwave power density, absolute pressure and thickness of apple slices) during the microwave vacuum drying processing into account, a model based on the orthogonal test analytic technique will be build. The function of this model is to analyses the effect of these variables and their interaction on the product quality, so that optimal processing parameters can be determined.

---

Detection of spoilage in chicken fillets using near-infrared chemical imaging and chemometrics (PhD)

Feng YZ, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Chemical imaging was exploited for its potential in direct and fast determination of total viable counts (TVC) in raw chicken breast fillets. Calibration models were established using chemical images acquired on a line-scan chemical imaging system (900-1700 nm) as well as chemometric methods including spectral transforms and partial least squares regression. Prior to modelling, images were transformed to different units, e.g. reflectance, absorbance and Kubelka-Munck (K-M). The best full-wavelength calibration model was obtained based on absorbance spectra, where the correlation coefficients (R) were 0.97 and 0.93, and the root mean squared errors (RMSEs) were 0.37 and 0.57 log10 colony forming unit (CFU) g-1 for calibration and cross validation, respectively. Simplified models were also built and the one based on K-M spectra was found to be optimum, with R and RMSEs of 0.96 and 0.40 log10 CFU g-1 as well as 0.94 and 0.50 log10 CFU g-1 for calibration and cross validation, respectively. It was pointed out that the prediction maps offered a novel way for visualizing the gradient of TVC on meat surface. Moreover, visualization maps could be used to investigate the adaptability of a developed calibration model.

---

Rapid and non-destructive determination of drip loss in salmon fillets using hyperspectral imaging technique (PhD)

He HJ, Wu D, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Drip loss is an important index in salmon quality evaluation. This work was carried out for rapid and non-destructive determination of drip loss distribution in salmon fillets using visible hyperspectral imaging (400-1000 nm). Hyperspectral images were acquired for salmon fillet samples and their spectral signatures were extracted. Partial least-squares regression (PLSR) was used to correlate spectral data of fillet samples with its reference drip loss value estimated by using a weight loss method. Eleven important wavelengths were selected by using the regression coefficients method to build an optimised PLSR model, leading to a correlation coefficient (RCV) of 0.834 and root-mean-square error estimated by cross-validation (RMSECV) of 0.067. The results showed that hyperspectral imaging combined with PLSR analysis offers an effective capability for determining the spatial distribution of drip loss in salmon fillets.

---