Automatic identification framework of the geometric parameters on self-piercing riveting cross-section using deep learning

Corrugated board, widely used in the packing industry, is a recyclable and durable material. Its strength and cushioning, influenced by geometry, environmental conditions like humidity and temperature, and paper quality, make it versatile. Double-walled (or five-ply) corrugated board, comprising two flutes and three liners, enhances these properties. This study introduces a novel approach to analyze five-layered corrugated board, extending a previously published algorithm for single-walled boards. Our method focuses on measuring the layer and overall board thickness, flute height, and center lines of each layer. Through the integration of image processing and genetic algorithms, the research successfully developed an algorithm for precise geometric feature identification of double-walled boards. Images were recorded using a special device with a sophisticated camera and image sensor for detailed corrugated board cross-sections. Demonstrating high accuracy, the method only faced limitations with very deformed or damaged samples. This research contributes significantly to quality control in the packaging industry and paves the way for further automated material analysis using advanced machine learning and image sensors. It emphasizes the importance of sample quality and suggests areas for algorithm refinement in order to enhance robustness and accuracy.

Section: Introductionmentioning

confidence: 99%

Deciphering Double-Walled Corrugated Board Geometry Using Image Analysis and Genetic Algorithms

Rogalka,

Grabski,

Garbowski

2024

“…Wyder and Lipson identified the static and dynamic properties of cantilever beams using the CNNs, basing their classification on raw cross-section images [ 27 ]. Li et al explored different deep learning techniques to analyze the geometric features of self-piercing riveting cross-section, with SOLOv2 and U-Net architectures yielding the best results [ 28 ]. Ma et al conducted a study on the geometrical parameters of crushed thin-walled carbon fiber-reinforced polymer tubes cross-sections [ 29 ].…”

Section: Introductionmentioning

confidence: 99%

In-Situ Classification of Highly Deformed Corrugated Board Using Convolution Neural Networks

Rogalka,

Grabski,

Garbowski

2024

The extensive use of corrugated board in the packaging industry is attributed to its excellent cushioning, mechanical properties, and environmental benefits like recyclability and biodegradability. The integrity of corrugated board depends on various factors, including its geometric design, paper quality, the number of layers, and environmental conditions such as humidity and temperature. This study introduces an innovative application of convolutional neural networks (CNNs) for analyzing and classifying images of corrugated boards, particularly those with deformations. For this purpose, a special device with advanced imaging capabilities, including a high-resolution camera and image sensor, was developed and used to acquire detailed cross-section images of the corrugated boards. The samples of seven types of corrugated board were studied. The proposed approach involves optimizing CNNs to enhance their classification performance. Despite challenges posed by deformed samples, the methodology demonstrates high accuracy in most cases, though a few samples posed recognition difficulties. The findings of this research are significant for the packaging industry, offering a sophisticated method for quality control and defect detection in corrugated board production. The best classification accuracy obtained achieved more than 99%. This could lead to improved product quality and reduced waste. Additionally, this study paves the way for future research on applying machine learning for material quality assessment, which could have broader implications beyond the packaging sector.

“…Wyder and Lipson examined convolutional neural networks for the identification of the static and dynamic properties of cantilever beams based on their raw cross-section images [18]. Li et al applied various deep learning techniques for analyzing the geometric features of a self-piercing riveting cross-section [19]. They showed that the SOLOv2 and Unet architectures give the best results.…”

Section: Introductionmentioning

confidence: 99%

Identification of Geometric Features of the Corrugated Board Using Images and Genetic Algorithm

Rogalka,

Grabski,

Garbowski

2023

The corrugated board is a versatile and durable material that is widely used in the packaging industry. Its unique structure provides strength and cushioning, while its recyclability and bio-degradability make it an environmentally friendly option. The strength of the corrugated board depends on many factors, including the type of individual papers on flat and corrugated layers, the geometry of the flute, temperature, humidity, etc. This paper presents a new approach to the analysis of the geometric features of corrugated boards. The experimental set used in the work and the created software are characterized by high reliability and precision of measurement thanks to the use of an identification procedure based on image analysis and a genetic algorithm. In the applied procedure, the thickness of each layer, corrugated cardboard thickness, flute height and center line are calculated. In most cases, the proposed algorithm successfully approximated these parameters.