Food tray sealing fault detection using hyperspectral imaging and PCANet

Benouis, Mohamed; Medus, Leandro D.; Saban, Mohamed; Łabiak, Grzegorz; Muñoz, Andrés Pedreño

doi:10.1016/j.ifacol.2020.12.1955

Cited by 8 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, it leads to enforce the output of hidden layer activation to be equal to zero. As a result, the data in the new feature space can be given as in Equation (27).…”

Section: Deep Learning Methods For Food Tray Classification Using Fused Data From Hyperspectral Imagesmentioning

confidence: 99%

See 1 more Smart Citation

Food Tray Sealing Fault Detection in Multi-Spectral Images Using Data Fusion and Deep Learning Techniques

et al. 2021

Self Cite

View full text Add to dashboard Cite

A correct food tray sealing is required to preserve food properties and safety for consumers. Traditional food packaging inspections are made by human operators to detect seal defects. Recent advances in the field of food inspection have been related to the use of hyperspectral imaging technology and automated vision-based inspection systems. A deep learning-based approach for food tray sealing fault detection using hyperspectral images is described. Several pixel-based image fusion methods are proposed to obtain 2D images from the 3D hyperspectral image datacube, which feeds the deep learning (DL) algorithms. Instead of considering all spectral bands in region of interest around a contaminated or faulty seal area, only relevant bands are selected using data fusion. These techniques greatly improve the computation time while maintaining a high classification ratio, showing that the fused image contains enough information for checking a food tray sealing state (faulty or normal), avoiding feeding a large image datacube to the DL algorithms. Additionally, the proposed DL algorithms do not require any prior handcraft approach, i.e., no manual tuning of the parameters in the algorithms are required since the training process adjusts the algorithm. The experimental results, validated using an industrial dataset for food trays, along with different deep learning methods, demonstrate the effectiveness of the proposed approach. In the studied dataset, an accuracy of 88.7%, 88.3%, 89.3%, and 90.1% was achieved for Deep Belief Network (DBN), Extreme Learning Machine (ELM), Stacked Auto Encoder (SAE), and Convolutional Neural Network (CNN), respectively.

show abstract

“…Thus, it leads to enforce the output of hidden layer activation to be equal to zero. As a result, the data in the new feature space can be given as in Equation (27).…”

Section: Deep Learning Methods For Food Tray Classification Using Fused Data From Hyperspectral Imagesmentioning

confidence: 99%

“…In our case, our first contribution is a novel hyperspectral food inspection algorithm based on a PCANet network [27]. The proposed machine learning algorithm detects any anomaly located in the seal by analyzing the mean value of spectral bands in the datacube.…”

Section: Introductionmentioning

confidence: 99%

Food Tray Sealing Fault Detection in Multi-Spectral Images Using Data Fusion and Deep Learning Techniques

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Table 1 presents summary of the equipment and product' developed by various AI technologies and their domains. Learning models are constructed to follow and forecast several environmental effects such as climate variation during crop production [8,9] Supply chain quality data integration method -Blockchain technology Logistics of agriculture products raising water availability [12,13] Product sorting/packaging -Sensor-based sorting system -Tensor flow ML-based system TOMRA [14,15] Fruit safety and quality -Gaussian Mixture Mode and IR vision sensor -Fourier Based separation model -Multi-resolution Wavelet transform and AI (classifier)of SVM and BPNN -FNN and SVM Smart refrigerator; Intelligent refrigerator [15][16][17] Food Quality ANN Forecast the quality loss as weight loss of frozen dough using ANN [18] Quality control -X-ray detection -MRI X-ray imaging detects defects and contaminants in agricultural commodities [19] Image processing -CNN -Hyperspectral imaging -PCANet Food tray packaging system; Food tray sealing fault detection [20,21] Forecasting of food production -Fuzzy logic -ML Predict the production and consumption of rice using ANN, SVM, GP, and GPR to predict future milk yield [22,23] Supply chain optimization -Evolutionary ML Scheduled transportation; reduced held inventory; cost in supply chain [24,25] Preparing and dispensing food -Robotics Food applications, drone and robotic deliveries, and autonomous cars [26] New food product development -ML -Deep learning algorithms Self-service soft drink corner [27] Identification of taste characteristics -Convolutional Neural Networks (CNN) -Multi-layer perceptron (MLP)-Descriptor -MLP Fingerprint MLP-Fingerprint model showed the best prediction results for bitterant/non-bitterant, sweetener/non-sweetener, and bitterant/sweetener [28] In precision agriculture, AI can be used to analyze data from sensors, drones, and satellites to optimize farming practices, such as irrigation, fertilization, and pest management. This can lead to higher yields, lower costs, and reduced environmental impact [29,30].…”

Section: Introductionmentioning

confidence: 99%

Artificial Intelligence: Implications for the Agri-Food Sector

et al. 2023

View full text Add to dashboard Cite

Artificial intelligence (AI) involves the development of algorithms and computational models that enable machines to process and analyze large amounts of data, identify patterns and relationships, and make predictions or decisions based on that analysis. AI has become increasingly pervasive across a wide range of industries and sectors, with healthcare, finance, transportation, manufacturing, retail, education, and agriculture are a few examples to mention. As AI technology continues to advance, it is expected to have an even greater impact on industries in the future. For instance, AI is being increasingly used in the agri-food sector to improve productivity, efficiency, and sustainability. It has the potential to revolutionize the agri-food sector in several ways, including but not limited to precision agriculture, crop monitoring, predictive analytics, supply chain optimization, food processing, quality control, personalized nutrition, and food safety. This review emphasizes how recent developments in AI technology have transformed the agri-food sector by improving efficiency, reducing waste, and enhancing food safety and quality, providing particular examples. Furthermore, the challenges, limitations, and future prospects of AI in the field of food and agriculture are summarized.

show abstract

“…The paper [1] presents a developed system in which a programmed algorithm for automatic detection of errors in the production of MPCG (Mobile Phone Cover Glass) is incorporated. In the food industry, in paper [2] photographs acquired by hyperspectral cameras are processed by software and used for automatic detection of errors in the food packaging process. In the field of mechanical engineering, the digital photo obtained during laser welding in paper [3] was processed by software and an algorithm was developed to monitor the accuracy of the product.…”

Section: Introductionmentioning

confidence: 99%

Problems of Software Processing of Digital Photography Intended for Determining the Dimensional Accuracy of a Product

Turinski¹,

Svalina²,

Lujić³

et al. 2023

Proceedings of the IMEKO TC11 Conference on Measurement in Testing, Inspection and Certification

View full text Add to dashboard Cite

Some of the problems that arise during the processing of the acquired photo are analyzed in this paper. The most commonly used software tools in image processing require binary photography, and its conversion into the binary form is required first. Conversion problems, and their proper solution, are key steps for further machine determination of dimensional accuracy. Removing smudges on binary photos and rotating displays for photo overlap methods are the problems that are solved in the work, in order to successfully prepare the photo with software for further determination of the dimensional accuracy of the product. The knowledge gained in this paper will help scientists, and facilitate their.

show abstract

Food tray sealing fault detection using hyperspectral imaging and PCANet

Cited by 8 publications

References 16 publications

Food Tray Sealing Fault Detection in Multi-Spectral Images Using Data Fusion and Deep Learning Techniques

Food Tray Sealing Fault Detection in Multi-Spectral Images Using Data Fusion and Deep Learning Techniques

Artificial Intelligence: Implications for the Agri-Food Sector

Problems of Software Processing of Digital Photography Intended for Determining the Dimensional Accuracy of a Product

Contact Info

Product

Resources

About