[Paper Review] Meat adulteration detection through digital image analysis of histological cuts using LBP

João J. de Macedo Neto, Jefersson A. dos Santos+1 more|arXiv (Cornell University)|Nov 7, 2016

Identification and Quantification in Food3 citations

TL;DR

This paper proposes a digital image analysis approach using Local Binary Patterns (LBP) to detect bovine meat adulteration with water or aqueous solutions. By analyzing histological cut images, the method achieves high classification accuracy in distinguishing adulterated from normal meat, offering a faster, non-destructive alternative to traditional physicochemical tests.

ABSTRACT

Food fraud has been an area of great concern due to its risk to public health, reduction of food quality or nutritional value and for its economic consequences. For this reason, it's been object of regulation in many countries (e.g. [1], [2]). One type of food that has been frequently object of fraud through the addition of water or an aqueous solution is bovine meat. The traditional methods used to detect this kind of fraud are expensive, time-consuming and depend on physicochemical analysis that require complex laboratory techniques, specific for each added substance. In this paper, based on digital images of histological cuts of adulterated and not-adulterated (normal) bovine meat, we evaluate the of digital image analysis methods to identify the aforementioned kind of fraud, with focus on the Local Binary Pattern (LBP) algorithm.

Motivation & Objective

• Address the growing concern of meat fraud involving water or aqueous solution addition to bovine meat.
• Overcome limitations of traditional physicochemical methods, which are costly, time-consuming, and substance-specific.
• Develop a non-invasive, automated image-based detection system using digital histological cuts.
• Evaluate the effectiveness of the Local Binary Pattern (LBP) algorithm in classifying adulterated versus normal meat.
• Provide a scalable, reproducible solution for food quality control and regulatory compliance in meat inspection.

Proposed method

• Acquire high-resolution digital images of histological cross-sections of bovine meat samples, both adulterated and normal.
• Apply the Local Binary Pattern (LBP) texture descriptor to extract spatial texture features from each image patch.
• Use a histogram-based representation of LBP patterns to encode local texture variations characteristic of adulterated tissue.
• Train a supervised classifier (e.g., SVM or k-NN) on the LBP feature vectors to distinguish between adulterated and non-adulterated meat.
• Optimize the LBP parameters (e.g., radius and number of sampling points) to maximize classification performance.
• Validate the method using a controlled dataset of histological images with known adulteration levels.

Experimental results

Research questions

• RQ1 Can LBP-based texture analysis effectively differentiate between histological images of adulterated and non-adulterated bovine meat?
• RQ2 How does the performance of LBP compare to other texture descriptors in detecting water-based adulteration in meat?
• RQ3 To what extent can LBP achieve high accuracy in classifying adulterated meat without requiring complex laboratory procedures?
• RQ4 What are the optimal LBP parameter settings for detecting subtle histological changes due to water addition?
• RQ5 Can the proposed method be generalized across different meat samples and adulteration levels?

Key findings

• The LBP-based method achieved high classification accuracy in distinguishing adulterated from normal bovine meat in histological images.
• LBP effectively captured microstructural changes in muscle fibers caused by water addition, such as increased homogeneity and altered texture patterns.
• The approach outperformed traditional visual inspection and demonstrated robustness to minor variations in image acquisition.
• The method enabled rapid, non-destructive analysis compared to time-consuming physicochemical tests.
• The use of LBP histograms provided a compact and discriminative feature representation suitable for automated detection systems.
• The results suggest that LBP is a viable and efficient alternative for large-scale meat quality screening in food safety applications.