International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 10 | Oct 2025
p-ISSN: 2395-0072
www.irjet.net
Early Detection of Chicken Breast Contamination with Salmonella Bacteria Using Artificial Intelligence (Deep Learning) Technique Niven Ahmad Alibraheem¹, Mohamad Shady Alrahhal² ¹*Faculty of Technical Engineering, Department of Food Engineering Technologies, Aleppo University, Syria
2AIMD Lab, College of Computing, Department of Computer Science, Fahad Bin Sultan University, Saudi Arabia
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Detecting Salmonella contamination in chicken
prediction related to existing of Salmonella contamination leads to providing dangerous food to individuals.
breast meat is a critical priority for the food industry, especially when it comes to talking about unseen infected regions of chicken breast meat. A novel intelligent system leveraging deep learning (ResNet) is introduced, incorporating Histogram Equalization (HE) to enhance training image quality. This preprocessing step improves feature extraction, enabling the model to identify previously undetectable regions of contamination. The system achieves 97% accuracy (with generalization) and 95% without, outperforming comparable methods by 1%. By integrating deep learning into food safety technologies, this approach can protect public health, boost consumer confidence, and enhance sustainability, particularly in developing nations like Syria, as well as in industrialized countries.
1.1 Statement of Problem Developing intelligent systems for detection of existing of Salmonella contamination manly depend on training the system on high-quality images. Therefore, good preprocessing step is critical to increase accuracy of prediction. Figure 1 illustrates the problem.
Key Words: Salmonella contamination, Artificial Intelligence, Deep Learning, Accuracy, Histogram Equalization.
1. INTRODUCTION On one hand, healthy chicken is a high-quality protein source, providing all essential amino acids needed for muscle growth, tissue repair, and immune function. It is also rich in B vitamins (B6, B12, niacin), which boost energy metabolism and support brain health. Additionally, chicken contains iron and zinc, enhancing oxygen transport and strengthening the body’s defence system, making it a powerful food for overall vitality [1]. On the other hand, contamination with Salmonella bacteria poses serious health risks, including severe food poisoning with symptoms like vomiting, diarrhea, fever, and abdominal cramps. In vulnerable groups (children, elderly, immunocompromised individuals), it can lead to life-threatening complications such as sepsis or meningitis [2]. Additionally, antibiotic-resistant Salmonella strains are emerging, making infections harder to treat and increasing public health concerns. Early detection of contamination with Salmonella is the best way to avoid such serious health risks.
Fig -1: Problem in terms of low accuracy. As shown in Figure 1, the correct prediction leads to high accuracy of prediction, which in turn leads to decrease the death ratio. On opposite, poor prediction leads to increase the death ratio.
1.2 Research Question The corresponding research question that is linked to the problem stated above is: how to increase detection of Salmonella contamination through providing effective technique of pre-processing linked with intelligent DL system. This research question can be solved by using histogram equalization technique to enhance the quality of images (Chicken Breast Contamination with Salmonella Bacteria), even if there is unseen amount of contamination by naked eyes, before training the intelligent system.
Artificial Intelligence (AI), particularly Convolutional Neural Networks (CNNs), has significantly advanced the detection of Salmonella contamination in food, offering faster, more accurate, and scalable solutions compared to traditional methods [3, 4]. However, poor accuracy of
© 2025, IRJET
|
Impact Factor value: 8.315
2. Literature Review (Past Studies) This section presents an overview of some related works that used artificial intelligence for detection of Salmonella contamination, as described below.
|
ISO 9001:2008 Certified Journal
|
Page 291