A novel method based on machine vision system and deep learning to detect fraud in turmeric powder

Convergent technologies to tackle challenges of modern food authentication

The authentication process involves all the supply chain stakeholders, and it is also adopted to verify food quality and safety. Food authentication tools are an essential part of traceability systems as they provide information on the credibility of origin, species/variety identity, geographical provenance, production entity. Moreover, these systems are useful to evaluate the effect of transformation processes, conservation strategies and the reliability of packaging and distribution flows on food quality and safety. In this manuscript, we identified the innovative characteristics of food authentication systems to respond to market challenges, such as the simplification, the high sensitivity, and the non-destructive ability during authentication procedures. We also discussed the potential of the current identification systems based on molecular markers (chemical, biochemical, genetic) and the effectiveness of new technologies with reference to the miniaturized systems offered by nanotechnologies, and computer vision systems linked to artificial intelligence processes. This overview emphasizes the importance of convergent technologies in food authentication, to support molecular markers with the technological innovation offered by emerging technologies derived from biotechnologies and informatics. The potential of these strategies was evaluated on real examples of high-value food products. Technological innovation can therefore strengthen the system of molecular markers to meet the current market needs; however, food production processes are in profound evolution. The food 3D-printing and the introduction of new raw materials open new challenges for food authentication and this will require both an update of the current regulatory framework, as well as the development and adoption of new analytical systems.

Denoising Diffusion Probabilistic Models and Transfer Learning for citrus disease diagnosis

Problems

Plant Disease diagnosis based on deep learning mechanisms has been extensively studied and applied. However, the complex and dynamic agricultural growth environment results in significant variations in the distribution of state samples, and the lack of sufficient real disease databases weakens the information carried by the samples, posing challenges for accurately training models.

Aim

This paper aims to test the feasibility and effectiveness of Denoising Diffusion Probabilistic Models (DDPM), Swin Transformer model, and Transfer Learning in diagnosing citrus diseases with a small sample.

Methods

Two training methods are proposed: The Method 1 employs the DDPM to generate synthetic images for data augmentation. The Swin Transformer model is then used for pre-training on the synthetic dataset produced by DDPM, followed by fine-tuning on the original citrus leaf images for disease classification through transfer learning. The Method 2 utilizes the pre-trained Swin Transformer model on the ImageNet dataset and fine-tunes it on the augmented dataset composed of the original and DDPM synthetic images.

Results and conclusion

The test results indicate that Method 1 achieved a validation accuracy of 96.3%, while Method 2 achieved a validation accuracy of 99.8%. Both methods effectively addressed the issue of model overfitting when dealing with a small dataset. Additionally, when compared with VGG16, EfficientNet, ShuffleNet, MobileNetV2, and DenseNet121 in citrus disease classification, the experimental results demonstrate the superiority of the proposed methods over existing approaches to a certain extent.

Assessment of cheese frauds, and relevant detection methods: A systematic review

Dairy products are widely consumed in the world due to their nutritional and functional characteristics. This group of food products are consumed by all age groups due to their health-giving properties. One of these products is cheese which has a high price compared to other dairy products. Because of this, it can be prone to fraud all over the world. Fraud in food products threatens the world's food safety and can cause serious damage to human health. There are many concerns among food authorities in the world about the fraud of food products. FDA, WHO, and the European Commission provide different legislations and definitions for fraud. The purpose of this review is to identify the most susceptible cheese type for fraud and effective methods for evaluating fraud in all types of cheeses. For this, we examined the Web of Science, Scopus, PubMed, and ScienceDirect databases. Mozzarella cheese had the largest share among all cheeses in terms of adulteration due to its many uses. Also, the methods used to evaluate different types of cheese frauds were PCR, Spectrometry, stable isotope, image analysis, electrophoretic, ELISA, sensors, sensory analysis, near-infrared and NMR. The methods that were most used in detecting fraud were PCR and spectrometry methods. Also, the least used method was sensory evaluation.

Review of visual analytics methods for food safety risks

With the availability of big data for food safety, more and more advanced data analysis methods are being applied to risk analysis and prewarning (RAPW). Visual analytics, which has emerged in recent years, integrates human and machine intelligence into the data analysis process in a visually interactive manner, helping researchers gain insights into large-scale data and providing new solutions for RAPW. This review presents the developments in visual analytics for food safety RAPW in the past decade. Firstly, the data sources, data characteristics, and analysis tasks in the food safety field are summarized. Then, data analysis methods for four types of analysis tasks: association analysis, risk assessment, risk prediction, and fraud identification, are reviewed. After that, the visualization and interaction techniques are reviewed for four types of characteristic data: multidimensional, hierarchical, associative, and spatial-temporal data. Finally, opportunities and challenges in this area are proposed, such as the visual analysis of multimodal food safety data, the application of artificial intelligence techniques in the visual analysis pipeline, etc.

Classification of Curcuma longa and Curcuma zanthorrhiza using transfer learning

Curcuma longa (turmeric) and Curcuma zanthorrhiza (temulawak) are members of the Zingiberaceae family that contain curcuminoids, essential oils, starch, protein, fat, cellulose, and minerals. The nutritional content proportion of turmeric is different from temulawak which implies differences in economic value. However, only a few people who understand herbal plants, can identify the difference between them. This study aims to build a model that can distinguish between the two species of Zingiberaceae based on the image captured from a mobile phone camera. A collection of images consisting of both types of rhizomes are used to build a model through a learning process using transfer learning, specifically pre-trained VGG-19 and Inception V3 with ImageNet weight. Experimental results show that the accuracy rates of the models to classify the rhizomes are 92.43% and 94.29%, consecutively. These achievements are quite promising to be used in various practical use.

Application of Analytical Technologies in the Discrimination and Authentication of Herbs from Fritillaria: A Review

Medicinal plants of Fritillaria are widely distributed in numerous countries around the world and possess excellent antitussive and expectorant effects. In particular, Fritillariae Bulbus (FB) as a precious traditional medicine has thousands of years of medical history in China. Herbs of Fritillaria have a high market value and demand while limited by harsh growing circumstances and scarce wild resources. As a consequence, fraudulent behaviors are regularly engaged by the unscrupulous merchants in an attempt to reap greater profits. It is of an urgent need to evaluate the quality of Fritillaria herbs and their products using various analytical instruments and techniques. This review has scrutinized approximately 160 articles from 1995 to 2022 published on the investigation of Fritillaria herbs and related herbal products. The botanical classification of genus Fritillaria, types of counterfeits, technologies applied for differentiating Fritillaria species were comprehensively summarized and discussed in the current review. Molecular and chromatographic identification were the dominant technologies in the authentication of Fritillaria herbs. Additionally, we brought some potential and promising technologies and analytical strategies into attention, which are worthy attempting in the future researches. This review could conduce to excellent reference value for further investigations of the authenticity assessment of Fritillaria species.

The Use of Convolutional Neural Networks and Digital Camera Images in Cataract Detection

Cataract is one of the major causes of blindness in the world. Its early detection and treatment could greatly reduce the risk of deterioration and blindness. Instruments commonly used to detect cataracts are slit lamps and fundus cameras, which are highly expensive and require domain knowledge. Thus, the problem is that the lack of professional ophthalmologists could result in the delay of cataract detection, where medical treatment is inevitable. Therefore, this study aimed to design a convolutional neural network (CNN) with digital camera images (CNNDCI) system to detect cataracts efficiently and effectively. The designed CNNDCI system can perform the cataract identification process accurately in a user-friendly manner using smartphones to collect digital images. In addition, the existing numerical results provided by the literature were used to demonstrate the performance of the proposed CNNDCI system for cataract detection. Numerical results revealed that the designed CNNDCI system could identify cataracts effectively with satisfying accuracy. Thus, this study concluded that the presented CNNDCI architecture is a feasible and promising alternative for cataract detection.