Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Smart Cameras: Using render image based object detection models in a Selective Laser Sintering part sorting system
To automatize the sorting process of SLS printed parts, a smart camera system running Neural Network based object classification algorithms is to be developed and tested. Influences on the AM process chain are to be analyzed.
Keywords: Machine Learning, Deep Learning, Neural Networks, Smart Cameras, SLS, Additive Manufacturing, Digital Process Chain, Internet of Things, IoT, Automatization, New Technologies, Innovations, Renderings
Smart Cameras can cover a wide range of sensor tasks in industrial IoT (Internet of Things) applications. One specific task is the monitoring of manual assembly work. Hereby, with the use of Neural Network based object detection algorithms, subtasks can be recognized automatically and workers can be supported with information about realized, forgotten or upcoming work steps.
In the group Design for New Technologies (DfNT) of pd|z, we are researching solutions to enable industrial applications based on smart camera technologies together with industry partners.
Analyzing accelerators and hurdles for application development as well as limiting the expertise needed to set up a new application by simplifying the process is thereby crucial.
Smart Cameras can cover a wide range of sensor tasks in industrial IoT (Internet of Things) applications. One specific task is the monitoring of manual assembly work. Hereby, with the use of Neural Network based object detection algorithms, subtasks can be recognized automatically and workers can be supported with information about realized, forgotten or upcoming work steps. In the group Design for New Technologies (DfNT) of pd|z, we are researching solutions to enable industrial applications based on smart camera technologies together with industry partners. Analyzing accelerators and hurdles for application development as well as limiting the expertise needed to set up a new application by simplifying the process is thereby crucial.
Printing parts using Selective Laser Sintering (SLS) requires manual sorting of the parts after fabrication and matching them to orders withs post-processing instructions. A system using smart cameras with an object classification model could automate this process.
To accelerate and automatize the supervised learning process of new models, the goal is to use rendered images based on the CAD data of the parts to be printed to create labelled training data. Running the model on a smart camera should enable fast classification and sorting of printed parts.
Testing of the application can be realized inhouse at pd|z or together with an industrial partner. Based on the results, the implication of the smart camera application on the AM process chain shall be described.
The process of creating training images using Blender and training a CNN has been realized. Your task is to fit this process to the described application, embed it into a digital process chain and test it.
Printing parts using Selective Laser Sintering (SLS) requires manual sorting of the parts after fabrication and matching them to orders withs post-processing instructions. A system using smart cameras with an object classification model could automate this process. To accelerate and automatize the supervised learning process of new models, the goal is to use rendered images based on the CAD data of the parts to be printed to create labelled training data. Running the model on a smart camera should enable fast classification and sorting of printed parts. Testing of the application can be realized inhouse at pd|z or together with an industrial partner. Based on the results, the implication of the smart camera application on the AM process chain shall be described.
The process of creating training images using Blender and training a CNN has been realized. Your task is to fit this process to the described application, embed it into a digital process chain and test it.
You are highly interested in working with Neural Networks and creating real world applications Desired are interest and skills in programming (Python preferred) and CAD Experience with the training of machine learning algorithms such as YOLO will help You work independently and are eager to interact with an industrial partner You approach problems systematically and manage your projects methodically You are a team player, curious and come up with innovative ideas You are willing to integrate into a professional work environment
We focus on human-centred product development and regard the link between research and education as the key to excellence in training. We see ourselves as a partner for industry and promote the continuous transfer of knowledge through cooperation, as well as the training and further education of students and graduates to strengthen the competitiveness of mechanical engineering industry.
We focus on human-centred product development and regard the link between research and education as the key to excellence in training. We see ourselves as a partner for industry and promote the continuous transfer of knowledge through cooperation, as well as the training and further education of students and graduates to strengthen the competitiveness of mechanical engineering industry.