Evaluation of Canning Quality Traits in Black Beans (Phaseolus vulgaris L.) by Visible/Near-Infrared Spectroscopy

摘要: Black bean (Phaseolus vulgaris L.) processing pre- sents unique challenges because of discoloration, breakage, development undesirable textures, and off-flavors during canning storage. These quality issues strongly affect standards consumer acceptance for beans. In this research, visible near-infrared (Vis/NIR) reflectance data the spectral region 400-2,500 nm were acquired from intact dry beans predicting five traits, i.e., hydration coefficient (HC), visual appearance (APP) color (COL), washed drained (WDC), texture (TXT), using partial least squares regression (PLSR). A total 471 samples harvested canned in 2010, 2011, 2012 used analysis. PLSR models based on Vis/ NIR showed low predictive performance, as measured by correlation prediction (Rpred )f or APP (Rpred= 0.275-0.566) TXT (Rpred=0.270-0.681), but better results HC (Rpred=0.517-0.810), WDC (Rpred=0.420- 0.796), COL (Rpred<0.533-0.758). comparison, measurements a colorimeter consistently good predictions (Rpred=0.796- 0.907). spite relatively poor agreement among sensory panelists determined multirater Kappa analysis (Kfree 0.20 0.18 COL), linear discriminant model Vis/NIR was able to classify into two categories "acceptable" "unacceptable", panelists' ratings traits beans, with classification accuracies 72.6 % higher. While technique has potential assessing improvements sensing instrumentation are need- ed order meet application requirements.