SET POINT DETERMINATION FROM SENSORY EVALUATIONS FOR FOOD PROCESS CONTROL

Sensory evaluation is often the ultimate measure of food quality, but food process control relies on instrumental measurements. Effective techniques are needed to convert desired sensory quality targets into instrumental process set points. This paper describes techniques developed for determining instrumental process set points from sensory evaluations. Various cases and different approaches depending on the nature of the sensory-instrumental relationships are outlined. the major issues addressed include additional constraints for underdetermined cases and reverse mapping with neural networks for nonlinear multivariate cases. These techniques were illustrated and tested with experimental data based on waffie samples. Seven sensory attributes were evaluated by trained panelists and instrumental measurements were obtained with a color computer vision system. For nonlinear multivariate cases, reverse mapping with neural networks successfully mapped sensory measurements to instrumental process set points with average errors less than 1.3%. the results demonstrate the effectiveness of the techniques developed.

[1]  Rosana G. Moreira.  Automatic Control for Food Processing Systems , 2001 .

[2]  Jinglu Tan,et al.  ACOUSTIC WAVE ANALYSIS FOR FOOD CRISPNESS EVALUATION , 1999 .

[3]  X. Gao,et al.  SEM IMAGE PROCESSING FOR FOOD STRUCTURE ANALYSIS , 1997 .

[4]  J. Tan,et al.  Classification of tough and tender beef by image texture analysis. , 2001, Meat science.

[5]  G Xie,et al.  Use of hyperbolic and neural network models in modelling quality changes of dry peas in long time cooking , 1999 .

[6]  H. Heymann,et al.  ANALYZING AND CORRELATING INSTRUMENTAL AND SENSORY DATA: A MULTWARIATE STUDY OF PHYSICAL PROPERTIES OF COOKED WHEAT NOODLES , 1999 .

[7]  Jinglu Tan,et al.  Evaluating expanded-food sensory properties by image analysis , 1999 .

[8]  J. G. Brennan,et al.  INTERRELATIONSHIPS BETWEEN SENSORY AND MECHANICAL CHARACTERISTICS OF CANNED PEACHES , 1994 .

[9]  Garmt Dijksterhuis,et al.  Procrustes Analysis in Studying Sensory‐Instrumental Relations , 1994 .

[10]  Ingemar Lundström,et al.  Crispbread quality evaluation based on fusion of information from the sensor analogies to the human olfactory, auditory and tactile senses , 1999 .

[11]  H. Heymann,et al.  Image texture features as indicators of beef tenderness. , 1999, Meat science.

[12]  Hans J. Rosenfeld,et al.  Prediction of sensory quality by near infrared reflectance analysis of frozen and freeze dried green peas (Pisum sativum) , 1990 .

[13]  X. Gao,et al.  APPLICATION OF FUZZY SETS AND NEURAL NETWORKS IN SENSORY ANALYSIS , 1999 .

[14]  J. Lu,et al.  Evaluation of pork color by using computer vision. , 2000, Meat science.

[15]  G. Hough,et al.  Partial least squares correlations between sensory and instrumental measurements of flavor and texture for Reggianito grating cheese , 1996 .