emabello42 / Traffic-Light-Detection-TensorFlow

Traffic Light Detection and Classification using TensorFlow

Traffic Light Detection and Classification using TensorFlow

---

Based on the work of coldKnight/TrafficLight_Detection-TensorFlowAPI.

---

Setup Environment

1. create conda environment conda create -n tf14 python=3.5 and activate it activate tf14
2. install tensorflow, for CPU pip install tensorflow==1.4.0 of for GPU

   conda install cudnn=6.0
   conda install cudatoolkit=8.0
   pip install tensorflow-gpu==1.4.0
   

3. clone tensorflow/models git clone https://github.com/tensorflow/models/ --branch r1.4.0 --single-branch and go to last commit that supports TensorFlow 1.4 git reset --hard 1f34fcafc1454e0d31ab4a6cc022102a54ac0f5b.
4. go into cd models/research and create protobuf files (if you have not installed protoc install it) by running protoc object_detection/protos/*.proto --python_out=. or something like “C:\Program Files\protoc-3.4.0-win32\bin\protoc.exe” object_detection/protos/*.proto --python_out=. on windows
5. install slim pip install -e slim/. and delf pip install -e delf/.
6. copy the object_detection folder to your workspace
7. install some dependencies pip install matplotlib==3.0.2 Pillow==5.4.1
8. and some more if you want to run the notebook to generate TFRecord files pip install jupyter==1.0.0 PyYAML==3.13 cython=0.29.4 opencv-python==4.0.0.21 tqdm==4.30.0, running pip install imgaug==0.2.7 might fail on windows, so you need to run conda install shapely=1.6.4 first

Simulator

Generate Train and Test Dataset

1. Download labeled images by coldKnight
2. copy sim_training_data folder into data/image_data/
3. run simulation_data.ipynb in data/ in order to generatesim_test.record and sim_train.record or download the TFRecord files we used

Setup Model for Training

1. select a model e.g. ssd_mobilenet_v1_coco and unpack it your workspace
2. download the corresponding configuration
3. update configuration (these are the settings we chose, trained witha GeForce GTX 1050 Ti ~10h of training)

num_classes: 90 => num_classes: 4
height: 300 => height: 400
width: 300 => width: 400
batch_size: 24 => 8
num_steps: 200000 => num_steps: 25000
max_detections_per_class: 100 => max_detections_per_class: 10
max_total_detections: 100 => max_total_detections: 10
fine_tune_checkpoint: "PATH_TO_BE_CONFIGURED/model.ckpt" =>fine_tune_checkpoint: "ssd_mobilenet_v1_coco_2017_11_17/model.ckpt"
PATH_TO_BE_CONFIGURED => data
num_examples: 8000 => num_examples: 974

remove data_augmentation_options as the dataset has already been augmented and max_evals: 10 as we want to test on all testdata.

Real World

Generate Train and Test Dataset

1. Download labeled images by coldKnight
2. copy real_training_data folder into data/image_data/
3. run real_data.ipynb in data/ in order to generatereal_test.record and real_train.record or download the TFRecord files we used

Setup Model for Training

1. reuse configuration from simulator
2. update PATH_TO_BE_CONFIGURED and num_examples

Train & Freeze

Train the Model

1. copy train.py and eval.py from object_detection in the root of you workspace
2. start training

   python train.py --logtostderr --train_dir=./trained_models --pipeline_config_path=ssd_mobilenet_v1_coco.config
   

3. while traing you can evaulate the model on the test dataset by running `

   python eval.py --logtostderr --pipeline_config_path=ssd_mobilenet_v1_coco.config --checkpoint_dir=trained_models --eval_dir=trained_models/eval/ --run_once=True
   

4. show training progress tensorboard --logdir=trained_models

Freezing the Graph

1. copy export_inference_graph.py from object_detection in the root of you workspace
2. freeze the graph

   python export_inference_graph.py --input_type image_tensor --pipeline_config_path ssd_mobilenet_v1_coco.config --trained_checkpoint_prefix trained_models/model.ckpt-25000 --output_directory frozen_graph
   

3. fix ValueError: Protocol message RewriterConfig has no "layout_optimizer" field. by editing line 72 in object_detection/exporter.py. Change layout_optimizer to optimize_tensor_layout.