Muhammad Lut, Liwauddin Abd Latib, M. A. Ayob, Nurasyeera Rohaziat
{"title":"FDM 3D打印中YOLOv5模型挤压下故障检测的比较","authors":"Muhammad Lut, Liwauddin Abd Latib, M. A. Ayob, Nurasyeera Rohaziat","doi":"10.1109/I2CACIS57635.2023.10193388","DOIUrl":null,"url":null,"abstract":"The fused deposited material (FDM) 3D printing technique has advanced to the point that it can now be utilised with a variety of materials. However, it still confronts challenges in quality control as flaws in the printed model frequently go undetected throughout the printing process. Consequently, time and materials easily go to waste. By employing an intelligent system that can detect faults throughout the printing process and pauses or halts the printing activity to apply corrective actions can prevent such issues. Hence, this paper presents a study on failure object detection of under extrusion in 3D printing by using YOLOv5 models of size n, s, m, l and xl with three different sets of data consisting of 600, 1200 and 2400 images. Simulation results showed that the YOLOv5 models were able to effectively detect under extrusion failures in 3D printing with varying accuracy based on the model size and data set. Based on the results, the YOLOv5xl model with a dataset of 2400 images achieved the highest detection accuracy among all models and datasets. These findings have potential implications for improving the reliability and efficiency of 3D printing process.","PeriodicalId":244595,"journal":{"name":"2023 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"YOLOv5 Models Comparison of Under Extrusion Failure Detection in FDM 3D Printing\",\"authors\":\"Muhammad Lut, Liwauddin Abd Latib, M. A. Ayob, Nurasyeera Rohaziat\",\"doi\":\"10.1109/I2CACIS57635.2023.10193388\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The fused deposited material (FDM) 3D printing technique has advanced to the point that it can now be utilised with a variety of materials. However, it still confronts challenges in quality control as flaws in the printed model frequently go undetected throughout the printing process. Consequently, time and materials easily go to waste. By employing an intelligent system that can detect faults throughout the printing process and pauses or halts the printing activity to apply corrective actions can prevent such issues. Hence, this paper presents a study on failure object detection of under extrusion in 3D printing by using YOLOv5 models of size n, s, m, l and xl with three different sets of data consisting of 600, 1200 and 2400 images. Simulation results showed that the YOLOv5 models were able to effectively detect under extrusion failures in 3D printing with varying accuracy based on the model size and data set. Based on the results, the YOLOv5xl model with a dataset of 2400 images achieved the highest detection accuracy among all models and datasets. These findings have potential implications for improving the reliability and efficiency of 3D printing process.\",\"PeriodicalId\":244595,\"journal\":{\"name\":\"2023 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS)\",\"volume\":\"48 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-06-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/I2CACIS57635.2023.10193388\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/I2CACIS57635.2023.10193388","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
YOLOv5 Models Comparison of Under Extrusion Failure Detection in FDM 3D Printing
The fused deposited material (FDM) 3D printing technique has advanced to the point that it can now be utilised with a variety of materials. However, it still confronts challenges in quality control as flaws in the printed model frequently go undetected throughout the printing process. Consequently, time and materials easily go to waste. By employing an intelligent system that can detect faults throughout the printing process and pauses or halts the printing activity to apply corrective actions can prevent such issues. Hence, this paper presents a study on failure object detection of under extrusion in 3D printing by using YOLOv5 models of size n, s, m, l and xl with three different sets of data consisting of 600, 1200 and 2400 images. Simulation results showed that the YOLOv5 models were able to effectively detect under extrusion failures in 3D printing with varying accuracy based on the model size and data set. Based on the results, the YOLOv5xl model with a dataset of 2400 images achieved the highest detection accuracy among all models and datasets. These findings have potential implications for improving the reliability and efficiency of 3D printing process.