Digital Image Filtering Optimization Supporting Iberian Ham Quality Prediction

Area of Science:

• Food Science and Technology
• Image Processing
• Quantitative Analysis

Background:

• Digital food images often exhibit variations in color and luminosity, necessitating image enhancement for accurate analysis.
• Multifractal analysis is a technique used to study complex patterns, such as fat infiltration in food products.
• Iberian ham quality is influenced by fat distribution and infiltration, requiring precise analytical methods.

Purpose of the Study:

• To compare the efficacy of high-pass filters versus no filtering for improving digital images of hand-cut Iberian ham.
• To assess the impact of image filtering on the accuracy of multifractal analysis for studying fat infiltration.
• To determine if enhanced image analysis can contribute to predictive techniques for Iberian ham quality.

Main Methods:

• Acquisition of digital images of hand-cut Iberian ham.
• Application of high-pass filters to a subset of the digital images.
• Conducting multifractal analysis on both filtered and unfiltered images to measure fractal dimensions.

Main Results:

• The use of high-pass filters resulted in more accurate fractal dimensions compared to unfiltered images.
• Filtered images provided a clearer representation of fat and its infiltration patterns within the Iberian ham.
• The improved accuracy in fractal dimension measurement has implications for quality assessment.

Conclusions:

• High-pass filtering is a crucial preprocessing step for enhancing the quality of digital food images used in multifractal analysis.
• Accurate fractal dimensions obtained through filtered images can significantly improve predictive models for Iberian ham quality.
• This approach offers potential for non-destructive quality control and product development in the cured meat industry.