Table A1.
Comparison of the proposals described in the state of the art.
| Ref. | Type | Application | Information Extracted From | AI Implemented | Year |
|---|---|---|---|---|---|
| [10] | App | Traffic jams | Smartphone sensors | Decision algorithm in function of the via status | 2018 |
| [11] | App | Traffic jams and incidents | Smartphone sensors and third parties | - | 2019 |
| [12] | App | Accident risk | Personal information and historical data | Fuzzy harmonic systems and fuzzy patterns | 2017 |
| [13] | App | Improve pedestrian routes | Open Street Maps and Google APIs | A * algorithm | 2018 |
| [14] | App | Improve pedestrian routes | Data base with architectural barriers and Google APIs | Optimizing iterative algorithm | 2020 |
| [15] | App | Improve pedestrian routes | Maps stored in Google Drive and Sensors | Dijkstra algorithm | 2020 |
| [16] | App | Improve pedestrian routes | Open Street Maps | Optimizing algorithm | 2018 |
| [17] | Camera on the road | Detect crossing intention | Cameras | Haarcascade based on OpenCV library; HOG based on SVM; SSD based on MobileNet; YOLO based on DNN | 2019 |
| [18] | Camera on the road | Detect crossing intention | Cameras | Region-based CNN; SVM; MLP | 2019 |
| [19] | Camera on the road | Detect crossing intention | Cameras | HOG based on SVM | 2017 |
| [20] | Camera on the road | Detect crossing intention | Cameras | KNN; SVM; ANN; DT; CNN | 2018 |
| [21] | Camera on the road | Detect crossing intention | Cameras | LSTM | 2020 |
| [22] | Camera on board vehicles | Detect crossing intention | Cameras | RF; SVM | 2017 |
| [23] | LIDAR sensor on the road | Detect crossing intention | LIDAR | DNN; LSTM; CNN | 2016 |
| [24] | Cameras and laser sensor on the road | Detect crossing intention | Cameras and laser sensors | AT-LSTM; SVM | 2020 |
| Proposed | App | Detect crossing intention and improve pedestrian routes | Google APIs, external data base and rotation vector | Fuzzy logic and optimizing algorithm | 2020 |
ANN: artificial neural network; AT-LSTM: long short-term memory network with attention mechanism; A* algorithm: A-Star search algorithm; CNN: convolutional neural network; DNN: dense neural network; DT: decision tree; HOG: histogram of oriented gradients; KNN: k-nearest neighbors; LSTM: long short-term memory; MLP: multilayer perceptron; RF: random forest; SSD: single shot detector; SVM: support vector machine; YOLO: you-only-look-once.