Deep-fried carbohydrate-rich foods items such as potato chips and French fries are one of the most popular snack foods consumed across the globe. In the production of these carbohydrate-rich foods items, a compound known as acrylamide is formed which is carcinogen and mutagen as well. The conventional chemical-based methods for detection of the presence of acrylamide in the deep-fried carbohydrate-rich food items are a time-consuming, destructive process that requires skilled manpower. The present work proposes a deep learning-based computer vision framework for automatic detection of the presence of acrylamide in potato chip samples with and without transfer learning. The performance of proposed six-layer CNN (without transfer learning) has been compared with the performance of the other transfer learned-models, for the present classification task using fivefold cross-validation. Experimental results show that the proposed six-layer CNN classifies the acrylamide-positive and negative samples with an average f1 score of 0.9251, whereas with the transfer learning-based approach, best average f1 score of 0.9644 was achieved. In conclusion, the proposed methodology in the current work is well suited for the acrylamide detection problem and the proposed work also analyses the effectiveness of the transfer learning-based approach when compared with the approach without utilizing the concept of transfer learning.

油炸富含碳水化合物的食品，例如薯片和炸薯条，是全球最受欢迎的休闲食品之一。在这些富含碳水化合物的食品的生产中，形成了一种称为丙烯酰胺的化合物，它也是致癌物和诱变剂。用于检测油炸的富含碳水化合物的食物中丙烯酰胺的存在的常规基于化学的方法是耗时，破坏性的过程，需要熟练的人力。本工作提出了一种基于深度学习的计算机视觉框架，可以自动检测带有或不带有转移学习的马铃薯片样品中丙烯酰胺的存在。拟议的六层CNN（无转移学习）的性能已与其他转移学习模型的性能进行了比较，使用五重交叉验证的当前分类任务。实验结果表明，提出的六层CNN对丙烯酰胺阳性和阴性样品进行了分类，平均f1得分为0.9251，而采用基于转移学习的方法，则获得了最佳平均f1得分为0.9644。总之，当前工作中提出的方法非常适合丙烯酰胺检测问题，并且与不使用转移学习概念的方法相比，所提出的工作还分析了基于转移学习的方法的有效性。最佳平均f1得分为0.9644。总之，当前工作中提出的方法非常适合丙烯酰胺检测问题，并且与不使用转移学习概念的方法相比，所提出的工作还分析了基于转移学习的方法的有效性。最佳平均f1得分为0.9644。总之，当前工作中提出的方法非常适合丙烯酰胺检测问题，并且与没有利用转移学习概念的方法相比，该方法还分析了基于转移学习的方法的有效性。