hh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Smart Kamera för Kvalitetskontroll av Rangebollar
Halmstad University, School of Information Technology.
Halmstad University, School of Information Technology.
2025 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

This project explores how quality control of golf balls can be automated using machine learning models. The purpose of automating this is to contribute to both a better environment and an improved playing experience. The models that have been trained and tested include CNN (Convolutional Neural Network), KNN (K-Nearest Neighbors), SVM (Support Vector Machine), and ViT (Vision Transformer). Experiments have also been conducted to isolate golf balls in order to simplify the classification task. In these experiments, YOLO (You Only Look Once), Mask R-CNN (Mask Region-Based Convolutional Neural Network), and ViT were tested. The results are not perfect, but the accuracy exceeded 50%. The conclusion is that more data is required to effectively train machine learning models than 1600 high quality images of 66 different unevenly distributed golfballs.

Abstract [sv]

Detta projekt går in på hur kvalitetskontroll av golfbollar kan automatiseras med hjälp av maskininlärningsmodeller, detta vill automatiseras då det kan bidra till en bättre miljö och spelupplevelse. De modeller som har tränats och prövats är CNN (Faltningsneuralnätverk), KNN (K-närmsta grannar), SVM (Stödvektormaskin) och ViT (Visiontransformator). Experiment gällande isolering av golfbollar för att förenkla problemet för klassificeringen har också skett, då har YOLO (Tittar Bara En Gång), Mask R-CNN (Mask regionbaserat faltningsneuralnätverk) och ViT testats. Resultaten är inte perfekta men noggranheten nådde över 50%. Slutsatsen är att det krävs mer data för att träna maskininlärningsmodeller än 1600 bilder av hög kvalitet på 66 olika ojämnt fördelade golfbollar.

Place, publisher, year, edition, pages
2025. , p. 69
Keywords [en]
AI, Machine learning, Quality control
Keywords [sv]
AI, Maskininlärning, Kvalitetskontroll
National Category
Computer Vision and Learning Systems Computer Engineering
Identifiers
URN: urn:nbn:se:hh:diva-56533OAI: oai:DiVA.org:hh-56533DiVA, id: diva2:1972362
External cooperation
Range Servant AB
Presentation
2025-05-23, 11:05 (Swedish)
Supervisors
Examiners
Available from: 2025-06-19 Created: 2025-06-18 Last updated: 2025-10-01Bibliographically approved

Open Access in DiVA

fulltext(2266 kB)77 downloads
File information
File name FULLTEXT02.pdfFile size 2266 kBChecksum SHA-512
b206807f8fc24eb7f5387917e12edfa70981ae4a67e797a032e0229d0fee107a5f516a72498dbb819340241b173966f211c47e060b2232f9f9dcffdf45b74ee2
Type fulltextMimetype application/pdf

By organisation
School of Information Technology
Computer Vision and Learning SystemsComputer Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 77 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 337 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf