Preliminary Study for Inspecting Moisture Content, Dry Matter Content, and Firmness Parameters of Two Date Cultivars Using an NIR Hyperspectral Imaging System.

The assessment and assurance of the quality attributes of dates is a key factor in increasing the competitiveness and consumer acceptance of this fruit. The increasing demand for date fruits requires a rapid and automated method for monitoring and analyzing the quality attributes of date fruits to replace the conventional methods used by inspection which limits the production and involves human errors. Moisture content (MC), dry matter content (DMC), and firmness (F) are three important quality attributes for two date cultivars (Khalas and Sukkari) that have been inspected using the hyperspectral imaging (HSI) technique based on the reflectance mode. Images of intact date fruits at the maturity stage Tamr were obtained within the wavelength range of 950-1750 nm. Monitoring and assessment of MC, DMC, and F [first maximum rupture force (MF, N)] were performed using a partial least squares regression model. Accurate prediction models were attained. The results highlight that the coefficients of determination (R2 Prediction) are estimated to be 0.91 and 0.89 for MC, DMC, and F (N) with the lowest values of the standard error of prediction (SEP) equal to 0.82, 0.81 (%), and 4.12 (N), respectively, and the residual predictive deviation (RPD) values were 3.65, 3.69, and 3.42 for MC, DMC, and F (N), respectively. The results obtained from this preliminary study indicate the great potential of applying HSI for the assessment of physical, chemical, and sensory quality attributes of date fruits overall in the five maturity stages.

Nondestructive evaluation of baking parameters on pogácsa texture.

This study aimed to investigate the potential application of image texture processing method on visible crumb structure of salty cake pogácsa, which was prepared with different baking times (5 and 7 min) and temperatures (200, 215, and 230°C). For this purpose, changes in gray level co-occurrence matrix (GLCM) features including energy, contrast, correlation, homogeneity, and entropy were monitored and their relationship with the instrumental texture parameters (hardness, adhesiveness, cohesiveness, springiness, gumminess, and chewiness) were assessed. The pore ratios were also extracted and visualized using image processing technique. Texture profile parameters indicated strong correlation (p < .01) with the image pattern parameters in different pogácsa groups. Gumminess showed strong correlation with contrast (0.503), correlation (-0.498), and homogeneity (0.401). Hardness also exhibited correlation with contrast (0.517), entropy (0.341), and correlation (-0.476). The pore ratio showed marked variation in crumb structure when different times and temperatures were used. Baking at 230°C for 7 min maximized the pore ratio (0.56). Penalty analysis revealed that oiliness, pore structure, and color of products were linked with baking time and temperature. Overall, the results suggested that the GLCM-based technique had the potential to be used as a nondestructive method for rapid quality assessment of pogácsa.

Methods for Improving Image Quality and Reducing Data Load of NIR Hyperspectral Images.

Near Infrared Hyperspectral Imaging (NIRHSI) is an emerging technology platform that integrates conventional imaging and spectroscopy to attain both spatial and spectral information from an object. Two important problems in NIRHSI are those of data load and unserviceable pixels in the NIR sensor. Hyperspectral imaging experiments generate large amounts of data (typically > 50 MB per image), which tend to overwhelm the memory capacity of conventional computer systems. This inhibits the utilisation of NIRHSI for routine online industrial application. In general, approximately 1% of pixels in NIR detectors are unserviceable or 'dead', containing no useful information. While this percentage of pixels is insignificant for single wavelength imaging, the problem is amplified in NIRHSI, where > 100 wavelength images are typically acquired. This paper describes an approach for reducing the data load of hyperspectral experiments by using sample-specific vector-to-scalar operators for real time feature extraction and a systematic procedure for compensating for 'dead' pixels in the NIR sensor. The feasibility of this approach was tested for prediction of moisture content in carrot tissue.