Subul Data Annotation

Subul Data Annotation

Subul Data Annotation

Subul Data Annotation

Case Study

Enhancing Product Shelf Monitoring with Precise OCR Annotation

OCR Product Text Detection and Classification

Platform

133 Refugees

in Syria impacted through this project

26,748 images

annotated polygons with ultra high precision

Polygon

Annotation Type

Classification

Shelf Monitoring

Challenge :

Our client faced significant challenges in accurately detecting and classifying the detailed text information printed on product labels. This included critical data such as production dates, ingredients, quality warnings, and storage instructions.

Inaccurate text recognition hindered the client’s ability to differentiate between various text types and organize products effectively. 

This misclassification caused inefficiencies in shelf management and product distribution.

Before 1After 1
Before2Smart warehouse management system using augmented reality techno

Solution

To address these challenges, Subul provided detailed and high-accuracy OCR annotations designed to enhance the client’s text detection model. By leveraging our expertise in image annotation, we ensured that every word and text category was identified and classified precisely. This level of detail improved the client’s ability to distinguish between diverse text types on product labels.

  • Specialized Annotation Team: Our team of 133 skilled workers, primarily from refugee backgrounds, brought a deep commitment to accuracy and a high standard of work ethics.
  • Advanced Annotation Techniques: We utilized Zebra’s polygon annotation tools to capture intricate text details, such as their locations and classifications, ensuring a complete and accurate dataset.
  • Custom Categorization: Each text element, including barcodes, expiry dates, and ingredient lists, was annotated with its corresponding category to streamline product identification and management.
  • Timely and Efficient Workflow: Our team adhered to strict timelines, ensuring the timely delivery of annotated data to support the client’s operations.

FEATURES

Implementation:

  • Specialized Team Deployment

    We assembled a dedicated team of 133 skilled workers from refugee backgrounds, specially trained in OCR annotation and text recognition. Their meticulous attention to detail was instrumental in achieving high accuracy.

  • Advanced Polygon Annotation

    Utilizing Zebra's advanced polygon annotation tools, we meticulously labeled the text types and their exact locations on product images. This included noting where texts appear and disappear and assigning appropriate classifications to each text segment.

  • Rigorous Quality Assurance

    Multiple rounds of thorough review and quality checks were conducted to maintain consistency and uphold the highest standards of annotation quality.

  • Timely Delivery

    Our team's commitment to efficiency ensured that we met all project deadlines, allowing the client's operations to proceed without interruption.

Before 3After 3

Result:

  • Improved Text Recognition: With the high-precision annotations provided by our team, the client significantly enhanced their ability to recognize various types of texts and their contents on product packaging. This includes clear and accurate identification of ingredients, expiration dates, and product-specific barcodes.
  • Precise Problem Identification: The accurate annotations enabled the client to pinpoint issues affecting products, such as expired items or ingredient discrepancies. This facilitated timely interventions to address such problems.
  • Optimized Shelf Distribution: By understanding the detailed textual information, the client could efficiently organize products on store shelves based on type similarity and material composition, correcting any errors in the distribution process.
  • Increased Compliance and Quality Assurance: The enhanced ability to monitor and classify product information ensured adherence to industry standards and improved overall product quality.

Conclusion

Through this project, Subul demonstrated how meticulous and high-quality OCR annotation can substantially impact product monitoring and quality assurance. Our team’s hard work and perfect execution not only met but exceeded the client’s expectations. By empowering the client with accurate and standardized data, we contributed to enhancing their operational efficiency and ensuring compliance with industry standards.

How Subul Data Annotation Works For You

Personalized and Fully Managed

Get a dedicated project manager to handle everything from guidelines to quality control, ensuring a seamless experience from start to finish.

Multilayer Quality Control

Experience the assurance of multi-tiered QC processes, including peer review and expert checks, guaranteeing a minimum accuracy of 98%.

Diverse and Scalable Workforce

Access a diverse team of industry experts, assembled within 72 hours, tailored to fit your project's specific needs and scale.

Unbiased and Compliant

Rely on our rigorously trained workforce for unbiased data, perfectly aligned with international privacy and AI standards.

Express Delivery

Benefit from our dynamic workforce, capable of rapidly scaling up to meet tight deadlines and ensure swift project delivery.

Direct Social Impact

Make a difference with every project; our entire workforce comes from refugee and disadvantaged backgrounds, contributing to social goals.

The world’s leading Ai teams Trust Us