Road Damage Detection and Classification with Detectron2 and Faster R-CNN

Pham, Vung; Pham, Chau; Dang, Tommy

doi:10.1109/bigdata50022.2020.9378027

Cited by 95 publications

(47 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model can be updated with future releases of YOLOv5, such as PP-YOLOv2 14 and tested with other networks such as Detectron 2 15 .…”

Section: Discussionmentioning

confidence: 99%

Pytri: A multi-weight detection system for biological entities

Mehdi

Lambert

2022

Preprint

View full text Add to dashboard Cite

The enumeration of biological entities is a critical part of experimental assays and usually requires large lengths of time. The standard method is to count the entities by hand or with OpenCV-based software, which can lead to inaccurate results. Here, we propose an online platform for biologists consisting of a system with multiple trained machine learning weights to detect various biological entities such as yeast colonies, bacterial colonies, and melanoma clusters. The Pytri model achieved a median relative error rate of 7.56% for bacterial and yeast colonies on Petri dishes, 6.58% for colonies on 96-well plates and 10.28% for melanoma cluster microscopy images. We showcase the application of state-of-the-art deep learning tools in bacterial entity detection, achieving significantly higher accuracy than traditional methods when compared to our base standard manual count.

show abstract

“…The model can be updated with future releases of YOLOv5, such as PP-YOLOv2 14 and tested with other networks such as Detectron 2 15 .…”

Section: Discussionmentioning

confidence: 99%

Pytri: A multi-weight detection system for biological entities

Mehdi

Lambert

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, computer vision models cannot be directly transferred to remote-sensing applications, in particular, due to the geospatial properties of the objects that appear in the scenes. As previously reported [25], the objects of interest in remote-sensing data are heterogeneous in terms of their size and shape and cover a wide range of spectral signatures depending on the sensor, scanning geometry, lighting, weather conditions, etc. Furthermore, compared with natural images (in which the focus is on the central foreground object of interest, the background is blurred, and images are collected at close range), remote-sensing images in general and UAS-based images, in particular, involve small objects (scanned from above from a given flight altitude) and complex backgrounds (where the surface of the ground around the object of interest does not have a specific focus or blurriness, and there is no prior identified or preferable foreground).…”

Section: Introductionmentioning

confidence: 94%

How the Small Object Detection via Machine Learning and UAS-Based Remote-Sensing Imagery Can Support the Achievement of SDG2: A Case Study of Vole Burrows

et al. 2021

View full text Add to dashboard Cite

Small mammals, and particularly rodents, are common inhabitants of farmlands, where they play key roles in the ecosystem, but when overabundant, they can be major pests, able to reduce crop production and farmers’ incomes, with tangible effects on the achievement of Sustainable Development Goals no 2 (SDG2, Zero Hunger) of the United Nations. Farmers do not currently have a standardized, accurate method of detecting the presence, abundance, and locations of rodents in their fields, and hence do not have environmentally efficient methods of rodent control able to promote sustainable agriculture oriented to reduce the environmental impacts of cultivation. New developments in unmanned aerial system (UAS) platforms and sensor technology facilitate cost-effective data collection through simultaneous multimodal data collection approaches at very high spatial resolutions in environmental and agricultural contexts. Object detection from remote-sensing images has been an active research topic over the last decade. With recent increases in computational resources and data availability, deep learning-based object detection methods are beginning to play an important role in advancing remote-sensing commercial and scientific applications. However, the performance of current detectors on various UAS-based datasets, including multimodal spatial and physical datasets, remains limited in terms of small object detection. In particular, the ability to quickly detect small objects from a large observed scene (at field scale) is still an open question. In this paper, we compare the efficiencies of applying one- and two-stage detector models to a single UAS-based image and a processed (via Pix4D mapper photogrammetric program) UAS-based orthophoto product to detect rodent burrows, for agriculture/environmental applications as to support farmer activities in the achievements of SDG2. Our results indicate that the use of multimodal data from low-cost UASs within a self-training YOLOv3 model can provide relatively accurate and robust detection for small objects (mAP of 0.86 and an F1-score of 93.39%), and can deliver valuable insights for field management with high spatial precision able to reduce the environmental costs of crop production in the direction of precision agriculture management.

show abstract

“…Detectron2 is the latest version of Detectron and is optimized with PyTorch. It has the advantages of flexibility and extensibility, so it is widely used in FAIR [ 10 , 11 , 12 ].…”

Section: Cnn Object Detection Based On Detectron2 Platformmentioning

confidence: 99%

“…Second, we select an optimal AI model that can simultaneously perform ship classification and structure recognition using visual sensors and repeatedly learned it using the MTMnet dataset. We compared and analyzed each model using the Detectron2 platform to apply diverse AI models [ 10 , 11 , 12 , 13 , 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

Application of Convolutional Neural Network (CNN) to Recognize Ship Structures

Lim

Kim

Hong

et al. 2022

Sensors

View full text Add to dashboard Cite

The purpose of this paper is to study the recognition of ships and their structures to improve the safety of drone operations engaged in shore-to-ship drone delivery service. This study has developed a system that can distinguish between ships and their structures by using a convolutional neural network (CNN). First, the dataset of the Marine Traffic Management Net is described and CNN’s object sensing based on the Detectron2 platform is discussed. There will also be a description of the experiment and performance. In addition, this study has been conducted based on actual drone delivery operations—the first air delivery service by drones in Korea.

show abstract

Road Damage Detection and Classification with Detectron2 and Faster R-CNN

Cited by 95 publications

References 21 publications

Pytri: A multi-weight detection system for biological entities

Pytri: A multi-weight detection system for biological entities

How the Small Object Detection via Machine Learning and UAS-Based Remote-Sensing Imagery Can Support the Achievement of SDG2: A Case Study of Vole Burrows

Application of Convolutional Neural Network (CNN) to Recognize Ship Structures

Contact Info

Product

Resources

About