[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119179-en":3,"doc-seo-119179-105":30,"detail-sidebar-cat-0-en-105":95},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119179,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Semmeldetector - Application of Machine Learning in Commercial Bakeries","Semmeldetector is a machine learning system that applies object detection to detect, classify, and count baked goods in images for commercial bakeries. The approach enables bakers to track unsold products, supporting production optimization and improved resource efficiency while meeting partner requirements. A dataset of 1151 images covering 18 baked-good types trains the models, with a Copy-Paste augmentation pipeline expanding limited data. YOLOv8 is trained and evaluated across data, model scale, and online augmentation settings, achieving up to 89.1% AP0.5.","Semmeldetector: Application of Machine Learning  \nin Commercial Bakeries  \nThomas H. Schmitt∗ , Maximilian Bundscherer and Tobias Bocklet  \nDepartment of Computer Science - Technische Hochschule Nu¨rnberg Georg Simon Ohm Nuremberg, Germany  \n∗ Email: [thomas.schmitt@th-nuernberg.de](thomas.schmitt@th-nuernberg.de)  \narXiv :2406 .04050v 1 [ cs .CV] 6 Jun 2024  \nAbstract—The Semmeldetector, is a machine learning application that utilizes object detection models to detect, classify and count baked goods in images. Our application allows commercial bakers to track unsold baked goods, which allows them to optimize production and increase resource efficiency. We compiled a dataset comprising 1151 images that distinguishes between 18 different types of baked goods to train our detection models. To facilitate model training, we used a Copy-Paste augmentation pipeline to expand our dataset. We trained the state-of-the-art object detection model YOLOv8 on our detection task. We tested the impact of different training data, model scale, and online image augmentation pipelines on model performance. Our overall best performing model, achieved an AP0.5 of 89. 1% on our test set. Based on our results, we conclude that machine learning can be a valuable tool even for unforeseen industries like bakeries, even with very limited datasets.  \nIndex Terms—machine learning, object detection, YOLOv8, image composition, baked goods, food inspection, industrial automation  \nI. INTRODUCTION  \nThe Semmeldetector, named after the locally used German word for bread bun, is a machine learning application that utilizes state-of-the-art object detection model YOLOv8 [11] to detect, classify, and count baked goods in images. However, due to the vast diversity of baked goods in Germany, with each bakery offering its unique assortment, to the best of our knowledge, there are no datasets available that sufficiently differentiate between baked goods. To train our models, we created a dataset comprising 1151 images distinguishing between 18 types of baked goods. We utilized SAM [13] to annotate our training data to streamline and speed up the annotation process. To facilitate model training, we employed a Copy-Paste augmentation [7] pipeline to expand our training data. Our object detection models allows commercial bakers to automatically track unsold baked goods, optimizing production, increasing resource efficiency, and meeting industry partner requirements. Which eliminates the often costly manual tracking that would otherwise be required. The main contributions of this study are: (1) The application of computer vision models in commercial bakeries to unsold product. (2) The demonstration of the effectiveness of the Copy-Paste augmentation [7] to enrich small datasets. (3) The deployment of our models as an iOS application, offering commercial bakeries a user-friendly platform to easily utilize our models.  \nThis study is partially supported by the European Social Funds (ESF) No. R.6-V0332 .2.43/1/5 .  \nA. Related Work  \nApplication studies [22] and [12] used U-net [19] and YOLOv5 [10] models to detect defects on or in baked goods, respectively. Both studies aimed at improving food safety using machine learning. The first utilized a combination of near infrared (NIR) spectroscopy images and computer vision to detect foreign contaminants in toast bread. The second used image data to detect mold on the surface of various food items, including baked goods. Both studies achieved detection accuracies of over 95% . These results demonstrate the effectiveness of machine learning in improving food safety.  \nApplication study [16] employed a SVM to perform image segmentation on images of a specific kind of flatbread. Their primary objective was to ensure quality control by accurately and quickly predicting the size and shape of bread sheets in various scenarios. To achieve this, they operated color-based in a relatively controlled image environment. They achieved a maxim","cbCaieUz6JQWWtL9","https://ap.wps.com/l/cbCaieUz6JQWWtL9","pdf",5254285,1,6,"English","en",105,"# Introduction\n## Related Work\n# Data\n## Training Set\n## Validation Set\n## Annotation and Augmentation\n# Model Training and Evaluation\n## Training Data Effects\n## Model Scale Effects\n## Online Image Augmentation\n# Results\n## Best Model Performance\n# Conclusion","[{\"question\":\"What does Semmeldetector do in commercial bakeries?\",\"answer\":\"Semmeldetector detects, classifies, and counts baked goods in images, helping bakeries track unsold products for better production planning and efficiency.\"},{\"question\":\"How large is the dataset and what does it cover?\",\"answer\":\"The study uses 1151 images that distinguish between 18 different types of baked goods, including visually similar items that are challenging even for human annotators.\"},{\"question\":\"Which model and training techniques were used?\",\"answer\":\"YOLOv8 is trained for the detection task, and a Copy-Paste augmentation pipeline expands the dataset to support model training with limited data.\"},{\"question\":\"What performance was achieved on the test set?\",\"answer\":\"The best model reaches an AP0.5 of 89.1% on the test set, indicating strong detection performance for the proposed workflow.\"}]","Semmeldetector - Application of Machine Learning in Commercial Bakeries | PDF",1785722945,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":90,"head_meta":92,"extra_data":94,"updated_unix":28},"semmeldetector-application-of-machine-learning-in-commercial-bakeries","",{"@graph":36,"@context":89},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/semmeldetector-application-of-machine-learning-in-commercial-bakeries/119179/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"What does Semmeldetector do in commercial bakeries?","Question",{"text":75,"@type":76},"Semmeldetector detects, classifies, and counts baked goods in images, helping bakeries track unsold products for better production planning and efficiency.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How large is the dataset and what does it cover?",{"text":80,"@type":76},"The study uses 1151 images that distinguish between 18 different types of baked goods, including visually similar items that are challenging even for human annotators.",{"name":82,"@type":73,"acceptedAnswer":83},"Which model and training techniques were used?",{"text":84,"@type":76},"YOLOv8 is trained for the detection task, and a Copy-Paste augmentation pipeline expands the dataset to support model training with limited data.",{"name":86,"@type":73,"acceptedAnswer":87},"What performance was achieved on the test set?",{"text":88,"@type":76},"The best model reaches an AP0.5 of 89.1% on the test set, indicating strong detection performance for the proposed workflow.","https://schema.org",{"og:url":52,"og:type":91,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":93,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":96},[97,101,105,109,114,118,123,126,131,134,138],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Exam",70,"exam",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},5,"Comic",60,"comic",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},"Technology",50,"technology",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":124,"slug":125},30,"research-report",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":129,"slug":130},9,"Religion & Spirituality",20,"religion-spirituality",{"id":129,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":129,"slug":133},"World Cup","world-cup",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":135,"slug":137},10,"Lifestyle","lifestyle",{"id":139,"doc_module":4,"doc_module_name":46,"category_name":140,"show_sort_weight":110,"slug":141},19,"General","general"]