Part 2: Detecting Truck Parking Lots on Satellite Images

This post describes a student group project developed within the Data Science Lab undergraduate course of the Vienna University of Economics and Business, co-supervised by Trustbit.

Student project team: Michael Fixl, Josef Hinterleitner, Felix Krause and Adrian Seiß

Supervisors:  Prof. Dr. Axel Polleres (WU Vienna), Dr. Vadim Savenkov (Trustbit)

Introduction

In the previous blog post (which you can find here), we created an already pretty powerful image segmentation model in order to detect the shape of truck parking lots on satellite images. However, we will now try to run the code on new hardware and get even better as well as more robust results.

New hardware and out-of-sample data

Up to now we trained all semantic segmentation models on Google Colab, which is advantageous as this assures everyone can reproduce the project easily but unfortunately the platform partly uses old package versions. In order to check out if the model training can simply be repeated on different hardware, we train the promising semantic segmentation models once again but with the newest package versions and on a high-end computer with an RTX 3090 graphics card. Hereby we can see that there can in fact be some randomness in the results depending on the hardware and packages used.

Let’s also check how robust our model is against more varied data, e.g. coming in different scales or from sources other than OpenStreetMap. Therefore, we made some experiments with unseen truck parking lots from Google Earth with various parking lot positions and sizes. Checking those images with the optimized PSPNet model from the last blog post shows that generalizability is not yet convincing, as you can see below.

Further testing and optimization

To counter the above mentioned issues, we should implement some fixes to reduce the risk of overfitting and check generalizability of the other models. Interestingly, further testing and optimization now shows that the LinkNet architecture turns out to be very precise and much more generalizable than PSPNet. It reaches a mIoU of up to 75% in the test set while also achieving high precision on completely new data with different structure than the train data.

You can find some predictions of this model on unseen parking lots from Google Earth below. Also this model of course fails under certain conditions and does not predict anything or predicts falsely as you can see in the right sample image. If you are interested in more detailed results, you might check out the comparison of the model performance before and after the previously mentioned changes in the table at the end of the blog post. There you can also find the training process of the final LinkNet model.

Conclusion

Overall, in this narrowly defined task the optimized LinkNet model shows decent accuracy and generalizability. Instead of sending expensive workers on site to measure coordinates of unrecorded truck parking lots, this model could be used to get a sometimes rough but at the same time very cheap estimation of the shapes of parking areas. Knowing these, the current activity of a truck can be derived based on GPS signals. This information could then be used to optimize transport operations.

Finally, a small app concludes the project, which is visible in the screenshot above. It simply takes coordinates as input and outputs a prediction mask using the optimized LinkNet model from before.

Additional material

Testing results

*new hardware, code optimization and changes (overfitting fix, newest package versions, etc.)

Training process of final LinkNet model

Blog 11/22/22

Part 1: Detecting Truck Parking Lots on Satellite Images

Real-time truck tracking is crucial in logistics: to enable accurate planning and provide reliable estimation of delivery times, operators build detailed profiles of loading stations, providing expected durations of truck loading and unloading, as well as resting times. Yet, how to derive an exact truck status based on mere GPS signals?

Blog 12/19/22

Creating a Cross-Domain Capable ML Pipeline

As classifying images into categories is a ubiquitous task occurring in various domains, a need for a machine learning pipeline which can accommodate for new categories is easy to justify. In particular, common general requirements are to filter out low-quality (blurred, low contrast etc.) images, and to speed up the learning of new categories if image quality is sufficient. In this blog post we compare several image classification models from the transfer learning perspective.

Headerbild zu Microsoft FinOps
Service

FinOps Microsoft - Microsoft Cloud Cost Optimization

TIMETOACT offers license management and compliance consulting in the Microsoft area, as well as technical consulting (e.g. on the use of Azure Services).

Network Performance Management
Kompetenz

Network Performance Management

With Network Performance Management you can monitor the performance of complex IT landscapes. This enables you to detect capacity bottlenecks, unexpected deviations from normal operation as well as faults at an early stage and to remedy them immediately.

Kompetenz 10/19/22

Digital Transformation with Atlassian Tools

At catworkx, we digitalize business processes for our customers on the basis of Atlassian tools such as Jira and CoAnd you can also benefit from the flexibility, performance and transparency of Altlas

Headerbild zu IBM Spectrum Protect
Technologie 5/20/21

IBM Spectrum Protect

Spectrum Protect is the leading data protection solution from the American industry leader IBM. As TIMETOACT, we take a holistic view of cyber security and focus on building resilient IT and OT.

Kompetenz

Years of Atlassian expertise

Based on Atlassian solutions, we support and advise companies on the path to digital and agile transformation.

Blog 7/14/21

Building and Publishing Design Systems | Part 2

Learn how to build and publish design systems effectively. Discover best practices for creating reusable components and enhancing UI consistency.

Blog 9/13/22

Introduction to Functional Programming in F# – Part 2

Explore functions, types, and modules in F#. Enhance your skills with practical examples and insights in this detailed guide.

Blog 11/14/23

Part 2: Data Analysis with powerful Python

Analyzing and visualizing data from a SQLite database in Python can be a powerful way to gain insights and present your findings. In Part 2 of this blog series, we will walk you through the steps to retrieve data from a SQLite database file named gold.db and display it in the form of a chart using Python. We'll use some essential tools and libraries for this task.

Standort 9/3/24

catworkx US

As an Atlassian Platinum and Enterprise Solution Partner, we look forward to welcoming you to the USA.

Blog 8/10/22

So, I wrote a book

Join me as I share the story of writing a book on F#. Discover the challenges, insights, and triumphs along the way.

HCL Notes Domino
Technologie

Notes / Domino

Within the TIMETOACT GROUP we have decades of experience with Notes and Domino. As HCL Business Partner we offer you a fast and uncomplicated access to the manufacturer.

catworkx on Tour - Rückblick auf ITCS Karriere-Messe in Hamburg
Blog

A look back at the ITCS career fair in Hamburg

catworkx on tour – in this section, we would like to give you an insight into the events and trade fairs that we organise, either on our own or together with our parent company, the TIMETOACT GROUP.

Headerbild zu IBM DB2
Technologie

IBM Db2

The IBM Db2database has been established on the market for many years as the leading data warehouse database in addition to its classic use in operations.

Blog 7/6/21

They promised it would be the next big thing!

Haven’t we all been there? We have all been promised by teachers, colleagues or public speakers that this or that was about to be the next big thing in tech that would change the world as we know it.

Logo RedHat
Technologie 7/2/20

RedHat

We are RedHat Advanced Partner. With RedHat as the market leader in Open Source IT solutions, we support our customers in actively designing and implementing their cloud journey.

Standort 11/8/22

catworkx Switzerland

As one of the largest Atlassian Platinum and Enterprise Solution Partners in the DACH region, we look forward to welcoming you to our office in Winterthur.

Standort 11/8/22

catworkx Austria

As one of the largest Atlassian Platinum and Enterprise Solution Partners in the DACH region, we look forward to welcoming you to our office in Vienna.

Standort 11/8/22

catworkx Germany

As one of the largest Atlassian Platinum and Enterprise Solution Partners in the DACH region, we look forward to welcoming you to our office in Hamburg.

Bleiben Sie mit dem TIMETOACT GROUP Newsletter auf dem Laufenden!