As previously reported by CosmeticsDesign, document quality management software company GlobalVision helps cosmetics and personal care product companies maintain accurate product lists and labels, record keeping demonstrating safety, and other critical reporting documents. We are continually working to innovate our methods. Recently, GlobalVision announced its latest innovation for industry manufacturers and suppliers by introducing AI-powered optical character recognition (OCR) capabilities to its Verify proofing software.
To learn more about the recently released technology and its potential impact on cosmetics and personal care product manufacturers and suppliers, CosmeticsDesign spoke to Julie Meredith, Chief Marketing Officer at GlobalVision, for insight. .
CDU: Can you give us a brief background about your professional experience, your company, and your relationship with the cosmetics and personal care products industry?
Julie Meredith (JM): GlobalVision is a software company that helps regulated industries with large-scale content proofing. Our main goal is to streamline your quality control and compliance review processes, saving your team time so they can focus on other important tasks and getting your products to market faster.
Over the years, GlobalVision has become a leader in calibration and quality inspection software around the world, working with great global brands such as Procter & Gamble, L’Oréal, Bath & Body Works, Yves Rocher, and Shiseido. Our software helps these companies quickly identify errors in regulated content and effectively reduce time to market.
Our cloud-based calibration and inspection software, Verify, was introduced in 2021 and has gained significant popularity, especially in regulated industries such as personal care and cosmetics. It provides a solution that accelerates the development of documentation and labeling assets by automatically and accurately identifying deviations at each stage of the asset lifecycle.
More recently, we have started leveraging artificial intelligence, primarily for optical character recognition (OCR) functionality, and continue to develop rapidly with quarterly updates that bring new features to our users four times a year.
CDU: Can you walk us through the process of developing and implementing the AI-powered optical character recognition (OCR) feature in the Verify proofreading program?
JM: AI-powered OCR capabilities have always been a key element in our development roadmap. Integrating this is a natural progression for Verify, allowing us to harness the power of AI within the framework of our technology solutions.
Demand for this feature was significant among customers in highly regulated industries such as cosmetics. They were looking for a solution that would allow them to inspect documents character by character, even in flattened format, as this is a critical feature for maintaining document compliance throughout their workflow.
Our development team utilizes a variety of technologies, parameters, and processing power to effectively enhance the character recognition capabilities of our algorithms by feeding them with relevant data, steadily increasing these capabilities over time. I made it possible to improve it. This process allows the algorithm to properly convert the characters on the document into a machine-readable format.
We initially introduced the OCR tool to only our major customers, giving them the option to adopt this feature and actively test its accuracy and speed across a wide variety of files across multiple languages. Throughout this phase, our machine learning framework accelerated the learning curve and powered his AI-powered OCR tool into an increasingly robust solution.
In parallel, the client provided valuable feedback and suggestions through the product team, allowing us to fine-tune the algorithm step by step and improve its strength and accuracy with each iteration.
With the introduction of OCR solutions, users in the cosmetics sector can now meticulously inspect character-by-character throughout label files, significantly shortening the document review process and ultimately speeding time to market. It will be significantly shortened.
CDU: How does this feature work?
J.M.: Optical character recognition is an innovative technology that converts printed, scanned, or handwritten text or images into machine-encoded text (also known as live, editable text). This allows computers to recognize, understand, and manipulate text from a variety of sources.
The main goal of Verify’s AI-powered OCR capabilities is to make text more accessible and editable, allowing users to extract valuable information from physical documents and images and put it into a digital, searchable format. It’s about being able to convert.
It is also important to note that OCR is an AI field that focuses on recognizing and extracting text from images without live text. Although OCR itself is a specific application within AI, it relies on various AI techniques and algorithms to perform tasks such as machine learning.
CDU: What pain points does this feature solve for manufacturers, suppliers, and brands of cosmetics and personal care products?
J.M.: Previously, there was no OCR calibration technology., Cosmetic and personal care product manufacturers, suppliers, brands, and regulatory teams were faced with the challenge of processing flattened documents with rasterized text and struggled with manual processes. This required the team to perform visual text extraction, manual proofreading, and manual data entry.
Without OCR technology, having to manually proofread documents, labels, packaging, and other critical content not only hinders revision team efficiency but also exposes you to potential errors and compliance risks. , compromising consumer and patient safety.
Especially in the cosmetics industry, imagine having to proofread allergy warnings and labels that aren’t in live text. OCR allows remediation teams to convert flattened text into live, editable text, making editing easier and ensuring complete accuracy and compliance of all critical content.
Other obstacles often arise when faced with the localization and translation tasks required for multinational operations. Global cosmetics and personal care manufacturers are particularly aware of this issue, as they must ensure accuracy and compliance of critical content in multiple different languages and scripts.
Verify’s latest AI-powered optical character recognition technology solves these problems and revolutionizes the way cosmetics industry regulatory, labeling, and promotion teams handle text within the document review process, providing automation and accuracy. Designed to bring sexuality to the forefront.
CDU: How is AI utilized in this feature?
J.M.: With Verify’s OCR feature, users can convert digital images into readable live text formats to easily edit documents such as labels, e-labels, screenshots, handwritten documents, images of text, and supplier proofs. You can inspect flattened text.
Verify’s OCR technology relies on machine learning, a subset of artificial intelligence (AI) technology. That is, it uses machine learning and computer vision algorithms to recognize characters and words in images and documents. This involves using computational techniques to perform tasks that typically require human intelligence or manual labor, such as reading and understanding text in images.
CDU: Does GlobalVision have any plans for further innovation and development using AI tools in label proofing technology?
J.M.: In the future, GlobalVision will continue to develop AI-powered tools designed to simplify and speed up label proofing.
One of our latest developments in this category is the incorporation of machine learning into the core technology that powers our spell checking algorithms. This allows you to review files from a human perspective, ignoring everything else and focusing only on the true and meaningful spelling mistakes.
This rapidly advancing “AI spell checking” represents a major advance for those involved in label proofing, as it will greatly increase the speed and accuracy of their work, and ultimately allow them to approve labels before they go to market. Masu.
A simple review of FDA recall records reveals that the majority of recalls are actually due to non-compliance issues with allergen statements. In the future, we may be able to harness the power of machine learning to teach Verify how to automatically recognize allergen descriptions on label files and flag them to users.
In this case, the platform recognizes the allergen description and analyzes it in the same way a human would do, ensuring compliance with FDA requirements. This simple addition to the platform can further automate regulatory compliance review processes and potentially reduce the risk of recalls by resolving non-compliance issues before they go to market.
