I’m building a road lane warning systems, using concepts of computer vision with modules such as OpenCV library and deep learning, a system that uses Machine learning to monitor lane markings and detect when a vehicle is moving out of its lane, a warning system (such as an audio, visual, vibration, or other alert) will notify the motorist of the unintended lane change so the motorist would be able to manoeuvre the vehicle back into its initial lane.
Steps done so far :
Data set selection and preparation
I have generated my own dataset (977 images)
I have pre-processed my data (resizing all the images using OpenCV 184 x 184 size ).
This project initially had the original dataset in the ratio of 100:19 as positive to negative class ratio. To balance them appropriately and not generate heavily duplicated data, only 19% of the positive class images have been augmented with Snow cover, Rain cover, Cloud cover and Fog. While, all the augmentations have been applied to all the images in the negative class images. After augmentation all the images have been resized to 184 x 184 size images to keep the image matrix for the deep learning small enough without losing the resolution of the lane markings. After the process was done it resulted in the dataset ratio of 1:1.
Steps pending :
This must include an introduction on the tools and infrastructure selected (cloud, Tensorflow etc.). This section should be a maximum of 1000 words plus visualizations and tables. This also must discuss the development of the model, from initial investigations to the final refined version
o Tool/infrastructure selection and rationale (could be multiple):
▪ Azure DSVM, AWS, WEKA
▪ Tensorflow, Scikit-learn, Tableau, PowerBI
o Model Development:
▪ ML algorithm selection
▪ ANN topology
▪ Selection criteria (10FCV, Loss, Accuracy, Sensitivity, Gridsearch)
Performance and Model Outcome
The results should present the results of the experiment/tool. The results should also include techniques used to validate the model/show that it would generalize. The selection of these techniques should also be detailed (with numerical values included). Usually, the findings should also be visualized and/or presented in a tabular manner. The performance should be briefly discussed here. This section should be a maximum of 1000 words plus visualizations and tables.
Because of the pressure, complexity, and lack of time I am finding it difficult to complete this project by myself.
Therefore, I am looking for assistance and willing to pay for it.
I look forward to hearing from you.
5 freelanceri licitează în medie 191$ pentru acest proiect
Alert: I will give you 20% discount on my bid rate also give on my All Services. So grabs this special offer is limited. Let�s get to the point. I am an highly experienced freelancer and an IIT Roorkee passout. I am Mai multe
hello sir, i am highly interested in your project. i have gone through your requirements and i believe i can be a valuable asset for your project. i am an expert in machine learning, deep learning and computer vision i Mai multe
Hi, I hope you are doing fine. I have almost 10 years of experience in machine learning algorithms. I can implement various types of artificial intelligence algorithms including yours with Matlab, Python and etc. I hav Mai multe
Hi, I have a good experience in machine learning and deep learning, including computer vision and openCV, so I'm ready to work with you to fulfill your requirements. Please contact me to discuss the details. Regards.
Hi, Greetings for the day! We are a team of dedicated and dynamic Machine Learning Engineers based in Mumbai, India, who have developed and deployed complex AI solutions using Machine Learning and Computer Vision. U Mai multe