Automated detection and classification of spilled loads on freeways based on improved YOLO network

Zhou, Siqi; Bi, Yi; Xu, Wei; Liu, Jiachen; Ye, Zixin; Li, Feng; Du, Yuchuan

doi:10.1007/s00138-021-01171-z

Cited by 19 publications

(8 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One augmentation technique in particular, known as copy-and-paste, has been used by several authors to expand DLOs segmentation datasets [25,26]. This technique consists of pasting objects (in this case cables), which have been segmented from real images, into different backgrounds, creating new images [27]. These approaches are fast and simple to implement, yet they have several drawbacks, such as requiring some manual work to capture and label the initial images, or not considering variations in features like lighting, shadows, reflections, or the shape of the DLOs, which reduces the realism of the generated images.…”

Section: Synthetic Image Segmentation Datasetsmentioning

confidence: 99%

Generation of realistic synthetic cable images to train Deep Learning segmentation models

Fresnillo,

Mohammed,

Vasudevan

et al. 2024

Preprint

View full text Add to dashboard Cite

One of the main contemporary challenges in robotics is the automation of tasks that involve handling Deformable Linear Objects (DLOs), such as cables or ropes. Due to their changing shape, constant perception is required, which makes computer vision, and in particular, image segmentation, fundamental. Even though image segmentation is a very studied problem, which has been addressed with high accuracy by many different Deep Learning models, they need to be properly trained to segment the objects of interest (DLOs in this case). To do this, it is necessary to have a large and diverse dataset. However, such a dataset hasn’t been created for cable images, and doing it manually would be complicated and extremely time-consuming. This paper addresses this issue with a novel methodology, which automatically generates synthetic cable datasets to train image segmentation models. This methodology utilizes Blender to create photo-realistic scenes and a Python pipeline to interact with them. To ensure the diversity of the dataset, before the generation of each image, the pipeline performs random variations on the elements of the scene. After this, the scene is animated, dropping the cables from a certain height and letting them deform naturally after landing, which results in realistic arrangements of the cables. The effectiveness of the methodology was demonstrated by training six popular segmentation models with synthetic datasets and using them to segment real cable images with great results (IoU over 70% and Dice coefficient over 80% for all the models).

show abstract

Section: Synthetic Image Segmentation Datasetsmentioning

confidence: 99%

Generation of realistic synthetic cable images to train Deep Learning segmentation models

Fresnillo,

Mohammed,

Vasudevan

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The YOLO framework has been successfully used in several diverse applications of civil engineering, such as pedestrian detection [46], real-time face detection [47], license plate detection [48], spilled load detection on freeways [49], pothole detection [50], traffic load distribution detection [51], worker and heavy construction equipment identification on site [52], building component identification [53], rebar diameter estimation [54], building footprint estimation [55], traffic management [56], pavement distress detection [57], crack detection [58][59][60][61], and maintenance [62][63][64]. The following sections describe the details of the structure of YOLOv5.…”

Section: Proposed Model For the Detection Of Cracks And Determination...mentioning

confidence: 99%

Two-stage method based on the you only look once framework and image segmentation for crack detection in concrete structures

Mishra

Jain

Singh

et al. 2022

Archit. Struct. Constr.

View full text Add to dashboard Cite

Detecting the presence of cracks and identifying their severity are crucial tasks for determining the structural health of a concrete building. In this study, we develop a two-stage automated method based on the You Only Look Once (YOLOv5) deep learning framework for the identification, localization, and quantification of cracks in the concrete structures. In the first stage, cracks are identified and localized using bounding boxes, while in the second stage, the length of cracks and, therefore, the damage severity are determined. The developed deep learning model is trained using 4500 annotated images from a total of 40000 images of size 227 × 227 pixel, which are obtained from an open-source dataset collected at various campus buildings of Middle East Technical University (METU). The concept of transfer learning (i.e., pre-trained weights) is used for the training, which drastically reduces the training time. The detection and localization accuracy of this model is measured in terms of the average precision, average recall, and F1-score. The YOLOv5 model achieves the mean average precision (mAP_0.5) of 95.02%. A ResNet model is also developed just to capture the supremacy of the YOLOv5 model. The proposed method can help in identifying structural anomalies through real-time monitoring that must be urgently repaired and thus can be used in high-quality civil infrastructure monitoring systems.

show abstract

“…While numerous methodologies have been proposed for path planning, selecting an approach that guarantees optimal paths remains a priority. Recent research incorporates neural networks, reinforcement learning, and SARSA algorithms to train robots effectively in path planning, ensuring collision-free trajectories (11)(12)(13)(14)(15) . The SARSA algorithm, in particular, stands out for achieving obstacle-free paths in unknown environments, making it a suitable choice for navigating hospital settings without collisions (16)(17)(18) .…”

Section: Introductionmentioning

confidence: 99%

Yolo Based Vision-Aided Obstacles Navigation and Avoidance with Path Planning using Sarsa Algorithm for Biped Robot in an Uncertain Hospital Environment

Duhan,

Panwar

2024

IJST

View full text Add to dashboard Cite

Background: This paper presents a method devised to ensure the secure navigation of biped robots within hospital environments by meticulously identifying an optimal collision-free path from initial to final positions while mitigating potential damage from undetected small objects. Methods: Implementation of a modified SARSA algorithm facilitates the seamless movement of biped robots amidst unknown environments replete with static obstacles. Extensive evaluation of several YOLO algorithms is conducted to ascertain accurate vision-based obstacle detection within hospital images. Subsequent utilization of the SARSA algorithm enables the planning of obstacle-free paths within the identified hospital setting. Findings: Within the evaluated YOLO algorithms, Yolov8 emerges as the pinnacle, showcasing unparalleled accuracy in object identification and refining bounding box precision for robot navigation within complex hospital environments. The SARSA-based path planning ensures the creation of collision-free routes, affirming safe traversal for the biped robot. Particularly noteworthy is Yolov8's exceptional precision in detecting minute objects, significantly reducing collision risks. Novelty: This research marks a significant stride in advancing human-like path planning for biped robots maneuvering through intricate hospital settings. The emphasis on accurate object identification stands as a linchpin for guaranteeing the robot's secure traversal. Significance: The integration of Yolov8 substantially augments the biped robot's capacity to precisely detect small objects, thereby mitigating collision risks and potential damage. Moreover, the successful application of the SARSA algorithm in planning obstacle-free paths within complex hospital environments holds promise for augmenting real-world robot navigation, especially in sensitive environments like hospitals.

show abstract

Automated detection and classification of spilled loads on freeways based on improved YOLO network

Cited by 19 publications

References 33 publications

Generation of realistic synthetic cable images to train Deep Learning segmentation models

Generation of realistic synthetic cable images to train Deep Learning segmentation models

Two-stage method based on the you only look once framework and image segmentation for crack detection in concrete structures

Yolo Based Vision-Aided Obstacles Navigation and Avoidance with Path Planning using Sarsa Algorithm for Biped Robot in an Uncertain Hospital Environment

Contact Info

Product

Resources

About