U-Infuse: Democratization of Customizable Deep Learning for Object Detection
Image data is one of the primary sources of ecological data used in biodiversity conservation and management worldwide. However, classifying and interpreting large numbers of images is time and resource expensive, particularly in the context of camera trapping. Deep learning models have been used to...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/8/2611 |
| _version_ | 1797538429315055616 |
|---|---|
| author | Andrew Shepley Greg Falzon Christopher Lawson Paul Meek Paul Kwan |
| author_facet | Andrew Shepley Greg Falzon Christopher Lawson Paul Meek Paul Kwan |
| author_sort | Andrew Shepley |
| collection | DOAJ |
| description | Image data is one of the primary sources of ecological data used in biodiversity conservation and management worldwide. However, classifying and interpreting large numbers of images is time and resource expensive, particularly in the context of camera trapping. Deep learning models have been used to achieve this task but are often not suited to specific applications due to their inability to generalise to new environments and inconsistent performance. Models need to be developed for specific species cohorts and environments, but the technical skills required to achieve this are a key barrier to the accessibility of this technology to ecologists. Thus, there is a strong need to democratize access to deep learning technologies by providing an easy-to-use software application allowing non-technical users to train custom object detectors. U-Infuse addresses this issue by providing ecologists with the ability to train customised models using publicly available images and/or their own images without specific technical expertise. Auto-annotation and annotation editing functionalities minimize the constraints of manually annotating and pre-processing large numbers of images. U-Infuse is a free and open-source software solution that supports both multiclass and single class training and object detection, allowing ecologists to access deep learning technologies usually only available to computer scientists, on their own device, customised for their application, without sharing intellectual property or sensitive data. It provides ecological practitioners with the ability to (i) easily achieve object detection within a user-friendly GUI, generating a species distribution report, and other useful statistics, (ii) custom train deep learning models using publicly available and custom training data, (iii) achieve supervised auto-annotation of images for further training, with the benefit of editing annotations to ensure quality datasets. Broad adoption of U-Infuse by ecological practitioners will improve ecological image analysis and processing by allowing significantly more image data to be processed with minimal expenditure of time and resources, particularly for camera trap images. Ease of training and use of transfer learning means domain-specific models can be trained rapidly, and frequently updated without the need for computer science expertise, or data sharing, protecting intellectual property and privacy. |
| first_indexed | 2024-03-10T12:31:21Z |
| format | Article |
| id | doaj.art-da55482e3ed14e1887d3af3ce5a481e2 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T12:31:21Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-da55482e3ed14e1887d3af3ce5a481e22023-11-21T14:39:53ZengMDPI AGSensors1424-82202021-04-01218261110.3390/s21082611U-Infuse: Democratization of Customizable Deep Learning for Object DetectionAndrew Shepley0Greg Falzon1Christopher Lawson2Paul Meek3Paul Kwan4School of Science and Technology, University of New England, Armidale, NSW 2350, AustraliaSchool of Science and Technology, University of New England, Armidale, NSW 2350, AustraliaSchool of Science and Technology, University of New England, Armidale, NSW 2350, AustraliaVertebrate Pest Research Unit, NSW Department of Primary Industries, P.O. Box 530, Coffs Harbour, NSW 2450, AustraliaSchool of IT and Engineering, Melbourne Institute of Technology, Melbourne, VIC 3000, AustraliaImage data is one of the primary sources of ecological data used in biodiversity conservation and management worldwide. However, classifying and interpreting large numbers of images is time and resource expensive, particularly in the context of camera trapping. Deep learning models have been used to achieve this task but are often not suited to specific applications due to their inability to generalise to new environments and inconsistent performance. Models need to be developed for specific species cohorts and environments, but the technical skills required to achieve this are a key barrier to the accessibility of this technology to ecologists. Thus, there is a strong need to democratize access to deep learning technologies by providing an easy-to-use software application allowing non-technical users to train custom object detectors. U-Infuse addresses this issue by providing ecologists with the ability to train customised models using publicly available images and/or their own images without specific technical expertise. Auto-annotation and annotation editing functionalities minimize the constraints of manually annotating and pre-processing large numbers of images. U-Infuse is a free and open-source software solution that supports both multiclass and single class training and object detection, allowing ecologists to access deep learning technologies usually only available to computer scientists, on their own device, customised for their application, without sharing intellectual property or sensitive data. It provides ecological practitioners with the ability to (i) easily achieve object detection within a user-friendly GUI, generating a species distribution report, and other useful statistics, (ii) custom train deep learning models using publicly available and custom training data, (iii) achieve supervised auto-annotation of images for further training, with the benefit of editing annotations to ensure quality datasets. Broad adoption of U-Infuse by ecological practitioners will improve ecological image analysis and processing by allowing significantly more image data to be processed with minimal expenditure of time and resources, particularly for camera trap images. Ease of training and use of transfer learning means domain-specific models can be trained rapidly, and frequently updated without the need for computer science expertise, or data sharing, protecting intellectual property and privacy.https://www.mdpi.com/1424-8220/21/8/2611animal identificationartificial intelligencecamera-trap imagescamera trappingdeep convolutional neural networksdeep learning |
| spellingShingle | Andrew Shepley Greg Falzon Christopher Lawson Paul Meek Paul Kwan U-Infuse: Democratization of Customizable Deep Learning for Object Detection Sensors animal identification artificial intelligence camera-trap images camera trapping deep convolutional neural networks deep learning |
| title | U-Infuse: Democratization of Customizable Deep Learning for Object Detection |
| title_full | U-Infuse: Democratization of Customizable Deep Learning for Object Detection |
| title_fullStr | U-Infuse: Democratization of Customizable Deep Learning for Object Detection |
| title_full_unstemmed | U-Infuse: Democratization of Customizable Deep Learning for Object Detection |
| title_short | U-Infuse: Democratization of Customizable Deep Learning for Object Detection |
| title_sort | u infuse democratization of customizable deep learning for object detection |
| topic | animal identification artificial intelligence camera-trap images camera trapping deep convolutional neural networks deep learning |
| url | https://www.mdpi.com/1424-8220/21/8/2611 |
| work_keys_str_mv | AT andrewshepley uinfusedemocratizationofcustomizabledeeplearningforobjectdetection AT gregfalzon uinfusedemocratizationofcustomizabledeeplearningforobjectdetection AT christopherlawson uinfusedemocratizationofcustomizabledeeplearningforobjectdetection AT paulmeek uinfusedemocratizationofcustomizabledeeplearningforobjectdetection AT paulkwan uinfusedemocratizationofcustomizabledeeplearningforobjectdetection |