Project information: Developement of integrated wood quality assesment system, collection of test data and deep learning improvement.
Project title: Developement of integrated wood quality assesment system, collection of test data and deep learning improvement.
Project number: No: 22.214.171.124/1
The aim of the project is to develop set of technologies to join information on wood collected in IR, NIR ,Visible light and X-Ray to identify defects and artefacts of the sawn wood to improve quality assessment and processing control.
About the Competence Center
Beneficiary: SIA “FARMĀCIJAS, BIOMEDICĪNAS UN MEDICĪNAS TEHNOLOĢIJU COMPETENCES CENTRS” intermediate body- CFLA; agreement no. 126.96.36.199/18/A/005 concluded on 29.04.2019.
The Pharmaceutical, Biomedical and Medical Technology Competence Center covers two scientific directions:
• Pharmacy, originator and generic medicines;
• Medicines of natural substances, food supplements, functional cosmetics.
The aim of the Center of Excellence is to support at least 11 researches in accordance with the defined research directions, developing new products and technologies in the sector and introducing them into production, thus promoting cooperation between the research and industry sector and increasing the competitiveness of the industry.
The project is implemented in the field of Latvia's smart specialization strategy - "Biomedicine, medical technologies, biopharmaceuticals and biotechnologies". As a result of the activities of the Center of Excellence, at least 11 research projects will be supported, at least 6 companies that will further introduce new products or technologies into production.
The total research costs of the project are 11,759,808.73 EUR, the planned amount of support from the European Regional Development Fund (hereinafter - ERDF) - 4,708,111.50 EUR.
The project is planned to be implemented by December 31, 2021
February 2, 2020
In the first phase, samples of ampoules with impurities and defects were examined on a rotating bench. The first prototype of a rotating stand has been developed and the first studies related to the visibility of defects in different lighting modes and camera modes have been performed. The next lighting and camera mode tests are set.
May 2, 2020
Technological schemes of the existing automatic and semi-automatic equipment have been studied.
Candidates for basic solutions selected:
01.01-02.05 - Adaptation of computer vision within the framework of image acquisition of semi-automatic equipment.
01.01-02.05 - Adaptation of computer vision by expanding semi-automatic equipment with a separate module (* currently the main candidate)
01.03-02.05 - A prototype of a mechanical system with a computer-controlled rotation model has been developed
15.04-02.05 -Received and evaluated sample images using the first lighting and camera candidates
The Obtained design requirements for the second prototype: Laboratory conditions have been prepared, equipment samples for the second round have been received and software for image acquisition and processing has been developed.
The first effective defect images were obtained, which to determine all the defects to be searched in the first approximation.
Selected lighting and camera mode technologies, as well as a common light and camera installation model that would allow the device to provide a stable, repeatable result
August 2, 2020
Automated prototype project created:
- technical design of the sample rotation system
- three-station model for lid, contents and glass control
- an improved laboratory prototype for testing has been created
- partially developed basic algorithms for product quality testing on test data
- a semi-automatic prototype project has been developed, based on an automatic prototype project, the aim of which is to test the obtained findings and model on the basis of one test station
February 2, 2021
Extended and built an automated prototype with additional functionality:
- Additional light solution appended to the prototype
-Tests of the new system started, data aquisition started
-Developed an annotation system. Allows to involve a large number of annotators working on building training and validation sets
-New type of cameras connected and integrated into the prototype.
A prototype of the second device has been created
The second prototype of the mechanical part of the equipment:
Obtained first sample images for Matrix Vision industrial cameras in a controlled environment with electronic and software sample rotation control
Lighting was created using:
backlight direct lighting Side lighting Software for the second prototype has been developed
Prototypes of industrial lighting have been ordered and received