A Very Deep Transfer Learning Model for Vehicle Damage Detection and Localization

Dhieb, Najmeddine; Ghazzai, Hakim; Besbes, Hichem; Massoud, Yehia

doi:10.1109/icm48031.2019.9021687

Cited by 32 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These losses are caused by inefficient claims processing, embezzlement, and bad business deliberation. With significant advancements in deep learning methods, these techniques have begun to be utilized in the insurance industry to address such challenges and mitigate their negative implications [13].…”

Section: Discussionmentioning

confidence: 99%

Vehicle Damaged Detection Using Deep Learning

2022

IRJMETS

View full text Add to dashboard Cite

Object Detection based Vehicle Damage Detection system can potentially save insurance companies Million. In this paper we propose a Vehicle Damage Detection System based on YOLO v4. We created a dataset of damaged vehicles and annotated regions consisting of Dents, Shattered Glass, damaged tail lights and scratches. We then pre-processed the images inline with the requirements of YOLO v4 and trained the model to achieve an mAP@50 of 81.20. The system proposed in this paper surpasses all previously proposed methods and gives promising results.

show abstract

Section: Discussionmentioning

confidence: 99%

Vehicle Damaged Detection Using Deep Learning

2022

IRJMETS

View full text Add to dashboard Cite

show abstract

“…The MAP was lower than 0.5 across all algorithms. Dhieb et al [28] used Inception-ResNetV2 to classify damage severity level, localize and detect part damage. Patil et al [29] and Dwivedi et al [30] used various CNN models to classify the car part damage, but these works only focused on a small set of car parts.…”

Section: Related Workmentioning

confidence: 99%

Evaluation of deep learning algorithms for semantic segmentation of car parts

Pasupa

Kittiworapanya

Hongngern

et al. 2021

Complex Intell. Syst.

View full text Add to dashboard Cite

Evaluation of car damages from an accident is one of the most important processes in the car insurance business. Currently, it still needs a manual examination of every basic part. It is expected that a smart device will be able to do this evaluation more efficiently in the future. In this study, we evaluated and compared five deep learning algorithms for semantic segmentation of car parts. The baseline reference algorithm was Mask R-CNN, and the other algorithms were HTC, CBNet, PANet, and GCNet. Runs of instance segmentation were conducted with those five algorithms. HTC with ResNet-50 was the best algorithm for instance segmentation on various kinds of cars such as sedans, trucks, and SUVs. It achieved a mean average precision at 55.2 on our original data set, that assigned different labels to the left and right sides and 59.1 when a single label was assigned to both sides. In addition, the models from every algorithm were tested for robustness, by running them on images of parts, in a real environment with various weather conditions, including snow, frost, fog and various lighting conditions. GCNet was the most robust; it achieved a mean performance under corruption, mPC = 35.2, and a relative degradation of performance on corrupted data, compared to clean data (rPC), of 64.4%, when left and right sides were assigned different labels, and mPC = 38.1 and rPC = $$69.6\%$$ 69.6 % when left- and right-side parts were considered the same part. The findings from this study may directly benefit developers of automated car damage evaluation system in their quest for the best design.

show abstract

“…In this work, recurrent neural network architecture was proposed in order to analyze individuals' payment history as well as predict their long-term possible social insurance payment behaviors. In [24], an automated deep-learning based architecture was proposed for vehicle damage detection and localization. This work suggests deep transfer and learning techniques for auto insurance companies to detect damage in vehicles, locate them, and classify their severity levels.…”

Section: Related Workmentioning

confidence: 99%

A Secure AI-Driven Architecture for Automated Insurance Systems: Fraud Detection and Risk Measurement

et al. 2020

Self Cite

View full text Add to dashboard Cite

The private insurance sector is recognized as one of the fastest-growing industries. This rapid growth has fueled incredible transformations over the past decade. Nowadays, there exist insurance products for most high-value assets such as vehicles, jewellery, health/life, and homes. Insurance companies are at the forefront in adopting cutting-edge operations, processes, and mathematical models to maximize profit whilst servicing their customers claims. Traditional methods that are exclusively based on human-in-theloop models are very time-consuming and inaccurate. In this paper, we develop a secure and automated insurance system framework that reduces human interaction, secures the insurance activities, alerts and informs about risky customers, detects fraudulent claims, and reduces monetary loss for the insurance sector. After presenting the blockchain-based framework to enable secure transactions and data sharing among different interacting agents within the insurance network, we propose to employ the extreme gradient boosting (XGBoost) machine learning algorithm for the aforementioned insurance services and compare its performances with those of other state-of-the-art algorithms. The obtained results reveal that, when applied to an auto insurance dataset, the XGboost achieves high performance gains compared to other existing learning algorithms. For instance, it reaches 7% higher accuracy compared to decision tree models when detecting fraudulent claims. The obtained results reveal that, when applied to an auto insurance dataset, the XGboost achieves high performance gains compared to other existing learning algorithms. For instance, it reaches 7% higher accuracy compared to decision tree models when detecting fraudulent claims. Furthermore, we propose an online learning solution to automatically deal with real-time updates of the insurance network and we show that it outperforms another online state-of-the-art algorithm. Finally, we combine the developed machine learning modules with the hyperledger fabric composer to implement and emulate the artificial intelligence and blockchain-based framework.

show abstract

A Very Deep Transfer Learning Model for Vehicle Damage Detection and Localization

Cited by 32 publications

References 9 publications

Vehicle Damaged Detection Using Deep Learning

Vehicle Damaged Detection Using Deep Learning

Evaluation of deep learning algorithms for semantic segmentation of car parts

A Secure AI-Driven Architecture for Automated Insurance Systems: Fraud Detection and Risk Measurement

Contact Info

Product

Resources

About