RoadCare: A Deep-learning Based Approach to Quantifying Road Surface Quality
Citations
312Β citations
36Β citations
11Β citations
8Β citations
6Β citations
References
802Β citations
"RoadCare: A Deep-learning Based App..." refers background in this paper
...We extract both raw as well as aggregated features from a single sample of 3 seconds and use them to train the SVM classifier....
[...]
...In this section we present the evaluation results for the SVM classifier in Section 4.1 and neural network based classification in Section 4.2....
[...]
...The evaluation of SVM with accelerometer features (ππππππ ) is shown in Section 4.1....
[...]
...We have evaluated the SVM classifier on our πΏπππ dataset for the set of features described in Section 3.2....
[...]
...Table 2 shows the classification accuracy for using SVM with βaccelerometer onlyβ features (ππππππ ) as well as with combined accelerometer and speed features (ππππππ+π ππππ ), for both linear and RBF (radial basis function) kernels....
[...]
630Β citations
"RoadCare: A Deep-learning Based App..." refers background in this paper
...Other similar works [47, 48] use deep neural networks to ascertain the presence of pavement distress in varying conditions....
[...]
480Β citations
"RoadCare: A Deep-learning Based App..." refers methods in this paper
...It first uses a watershed algorithm [4] to perform road segmentation, followed by Otsuβs binarization [39] and morphological thinning [5] to finally generate a topological skeleton of the pothole image....
[...]
450Β citations
"RoadCare: A Deep-learning Based App..." refers background in this paper
...Koch et al [23] uses geometric properties of a defect region, such as a pothole, and compares it with the texture of a non-defect region to detect if an anomaly is a pothole....
[...]
380Β citations
"RoadCare: A Deep-learning Based App..." refers methods in this paper
...We extract both raw as well as aggregated features from a single sample of 3 seconds and use them to train the SVM classifier....
[...]
...In this section we present the evaluation results for the SVM classifier in Section 4.1 and neural network based classification in Section 4.2....
[...]
...The evaluation of SVM with accelerometer features (ππππππ ) is shown in Section 4.1....
[...]
...We have evaluated the SVM classifier on our πΏπππ dataset for the set of features described in Section 3.2....
[...]
...Table 2 shows the classification accuracy for using SVM with βaccelerometer onlyβ features (ππππππ ) as well as with combined accelerometer and speed features (ππππππ+π ππππ ), for both linear and RBF (radial basis function) kernels....
[...]