Dr.-Ing.

Gruna, Robin

Filters

2 1
3
Reset filters

Settings
Now showing 1 - 3 of 3
  • Publication
    Multi-Modal Image Acquisition for AI-Based Bulky Waste Sorting (Incl. Terahertz Synthetic Aperture Radar)
    ( 2023)
    Cibiraite-Lukenskiene, Dovile
    ;
    Gundacker, Dominik
    ;
    Bihler, M.
    ;
    Heizmann, M.
    ;
    Schlüter, Friedrich
    ;
    Aderhold, Jochen
    ;
    Roming, Lukas
    ;
    Gruna, Robin
    ;
    Jonuscheit, Joachim
    ;
    Friederich, Fabian
    This work presents the results of the initial acquisition of a multi-modal dataset that will be utilized to train and test a neural network for wood sorting. The aim of the project is to improve wood recycling from bulky waste by using four complementary sensing systems: visual, infrared, terahertz, and thermography. The four systems were combined to capture 57 multi-modal images of bulky waste samples moving on the conveyor belt at a speed of 10 cm/s. Early fusion results on THz show 0.77 accuracy, whereas the best multi-modal data fusion accuracy equals 0.921.
  • Publication
    Multi-sensor data fusion using deep learning for bulky waste image classification
    ( 2023)
    Bihler, Manuel
    ;
    Roming, Lukas
    ;
    Jiang, Yifan
    ;
    Afifi, Ahmed J.
    ;
    Aderhold, Jochen
    ;
    Cibiraite-Lukenskiene, Dovile
    ;
    Lorenz, Sandra
    ;
    Gloaguen, Richard
    ;
    Gruna, Robin
    ;
    Heizmann, Michael
    Deep learning techniques are commonly utilized to tackle various computer vision problems, including recognition, segmentation, and classification from RGB images. With the availability of a diverse range of sensors, industry-specific datasets are acquired to address specific challenges. These collected datasets have varied modalities, indicating that the images possess distinct channel numbers and pixel values that have different interpretations. Implementing deep learning methods to attain optimal outcomes on such multimodal data is a complicated procedure. To enhance the performance of classification tasks in this scenario, one feasible approach is to employ a data fusion technique. Data fusion aims to use all the available information from all sensors and integrate them to obtain an optimal outcome. This paper investigates early fusion, intermediate fusion, and late fusion in deep learning models for bulky waste image classification. For training and evaluation of the models, a multimodal dataset is used. The dataset consists of RGB, hyperspectral Near Infrared (NIR), Thermography, and Terahertz images of bulky waste. The results of this work show that multimodal sensor fusion can enhance classification accuracy compared to a single-sensor approach for the used dataset. Hereby, late fusion performed the best with an accuracy of 0.921 compared to intermediate and early fusion, on our test data.
  • Publication
    Increasing the reuse of wood in bulky waste using artificial intelligence and imaging in the VIS, IR, and terahertz ranges
    ( 2023)
    Roming, Lukas
    ;
    Gruna, Robin
    ;
    Aderhold, Jochen
    ;
    Schlüter, Friedrich
    ;
    Cibiraite-Lukenskiene, Dovile
    ;
    Gundacker, Dominik
    ;
    Friedrich, Fabian
    ;
    Bihler, Manuel
    ;
    Heizmann, Michael
    Bulky waste contains valuable raw materials, especially wood, which accounts for around 50% of the volume. Sorting is very time-consuming in view of the volume and variety of bulky waste and is often still done manually. Therefore, only about half of the available wood is used as a material, while the rest is burned with unsorted waste. In order to improve the material recycling of wood from bulky waste, the project ASKIVIT aims to develop a solution for the automated sorting of bulky waste. For that, a multi-sensor approach is proposed including: (i) Conventional imaging in the visible spectral range; (ii) Nearinfrared hyperspectral imaging; (iii) Active heat flow thermography; (iv) Terahertz imaging. This paper presents a demonstrator used to obtain images with the aforementioned sensors. Differences between the imaging systems are discussed and promising results on common problems like painted materials or black plastic are presented. Besides that, pre-examinations show the importance of near-infrared hyperspectral imaging for the characterization of bulky waste.