AI-Enhanced Satellite Image Resolution

Challenge

The Current Situation

In the era of data-driven decision-making and artificial intelligence, organizations face significant challenges in acquiring, managing, and utilizing high-quality datasets. Among the main critical issues there is certainly the acquisition method and the maximization of the resolution of the data that will subsequently have to be processed by specific tools depending on the final use case.

Traditionally, the monitoring of large infrastructure has relied on manual visual inspections, involving both staff and suppliers, or occasionally through drone surveys or high-resolution satellite data. However, the current process requires significant manpower to collect, verify, process, and share information, leading to many hours of work for all the parties involved, generally including the need for a physical presence on site.

Furthermore, many large infrastructure projects span over extensive areas (often exceeding 300 hectares), which limits the feasibility of comprehensive on-site inspections. This often results in reduced frequency and coverage of spot surveys or necessitates longer working hours, making it almost impractical to survey the entire site using standard methods within a reasonable timeframe.

The Solution:

The goal of the challenge is to demonstrate the effectiveness of a super-resolution algorithm that can maximize the resolution of images for a specific area of interest, starting from freely available satellite data.

By leveraging advanced image processing techniques and machine learning, the algorithm should provide improved visual clarity and greater detail (x3 as a minimum upscaling factor), thereby minimizing the reliance on costly high-resolution datasets.

Enel aims to enhance the operational efficiency and the accuracy of assessments in scenarios such as construction site progress monitoring, with the aim of identifying structural components (i.e., wind turbines, PV modules, PV structures, earth movements, …);, identification of small/medium photovoltaic installations (i.e., covering 500-1.400m2); identification of abandoned / damaged / new buildings in of cities, etc.. The super-resolution images, produced by the winner of the challenge, will be used as input for computer vision and change detection algorithms, the latter being beyond the scope of this challenge. This will make it possible to support the afore-mentioned use cases, by tracking the evolution of conditions. This will enable more informed and timely decision-making through improved access to visual data.

The developed augmentation algorithm will need to be tested by the participants in this challenge by using reference geographical areas and infrastructure sites for which Enel has existing data available for comparison. These include, but are not limited to, Solar/Wind/BESS Power Plant construction sites, urban areas for detection of new photovoltaic installations and building shape & status identification.

The testing phase, which the challenge participants will conduct, will involve a detailed analysis of the algorithm's structure and performance by comparing the augmented outputs with the existing baseline data and based on the specific references and key performance indicators (KPIs) outlined below.

The aim is to ensure that the algorithm meets predefined standards of accuracy and reliability, providing valuable insights into its practical applicability for monitoring and analyzing large-scale infrastructure projects in the afore-mentioned use cases. 

The results of the test and results validation phase will enable the challenge winner to refine the algorithm, ensuring its robustness and effectiveness in diverse real-world Enel scenarios during a second phase project which isn’t part of the challenge process.

SOLUTIONS MUST HAVE:

TRL>6

Input data sources used: the algorithm must exclusively utilize free satellite data sources, such as Copernicus (i.e. Sentinel-2), Cosmo-skyMed, Prisma, Landsat or other similar publicly available datasets.

Test Areas: each participant can apply for more than one of the following Test areas (Click HERE for KML files for the test areas) 

1. Solar Power Plant Centurion, Italy, on 24 August 2023, 08 November 2023 and 23January 2024. Use case reference: Construction progress.
2. Province of Bologna, Italy, for the detection of new photovoltaic installations equal or larger than 500m2 in the time frame September 2024-December 2024.
3. Municipality of Bitonto, Italy, for abandoned / damaged / new Building recognition and classification on 11 April 2022 and 12 April 2022
4. A Wind Farm under construction, for Construction progress in Impofu (South Africa), for development status since March 2024.
5. BESS (Battery Energy Storage System) farm at the Pian di Giorgio under construction, for Construction progress on 07November 2024

The solution must be validated by the Participant by applying the following “KPI for validation”:

1. PSNR (Peak Signal-to-Noise Ratio): Measures the ratio between the signal and noise, indicating the quality of the reconstructed image compared to the original. A higher PSNR value indicates better image quality.
2. SSIM (Structural Similarity Index Measure): Evaluates the structural similarity between the original and enhanced images. It considers brightness, contrast, and structure, thereby providing an indication of perceived visual quality.
3. RMSE (Root Mean Square Error): Measures the root mean square error between the enhanced image and the reference image. A lower RMSE value indicates higher accuracy.
4. Processing Time: Measures the time taken by the algorithm to process the image. Efficient software should be able to handle large volumes of data in a reasonable timeframe.
5. Resource Utilization (CPU/GPU): Evaluates the efficiency of the algorithm in terms of memory usage and computational power.
6. Spectral Similarity Index: Measures how well the enhanced image retains the spectral data compared to the reference image.
7. Testing Under Different Environmental Conditions: Evaluates the algorithm's performance on images captured under various conditions (e.g., cloud cover, different seasons) in order to ensure it can be applied in different scenarios.
8. Testing with Different Resolutions and Satellite Datasets: The software should be tested on data from various satellites (e.g., Sentinel-2, Landsat) and different resolutions in order to assess its versatility.
9. LPIPS (Learned Perceptual Image Patch Similarity): A deep-learning metric that evaluates perceptual similarity using maps from neural networks.
10. Number of Use Cases treated successfully according to the afore-mentioned KPIs.

The participant must guarantee that any proposed solution featuring AI technology is compliant with EU and Italian law on AI and with the Regulations of the Challenge.

In order to accurately evaluate the performance of the algorithm, all the necessary data for calculating the previously mentioned KPIs must be provided at the end of the testing phase. This includes, for example, information such as the number of pixels, bit depth, image dimensions, and other relevant metadata.

Deliverables

Proposals must be submitted to the openinnovability.com platform in a single stage, and must include the following information in English:

• Solution name/title and overview
• A detailed description of the solution and model used
• Demo/Test of at least 5 images in the selected Use Cases- max file upload is 35MB- please provide link so that the evaluation team can access and verify the demo (please pay careful attention to the validity period of the link)
• For each processed image, the corresponding starting source image from the terrestrial observation service used must be provided, including its metadata
• The calculated KPIs for the images and the method of calculation
• A development roadmap to market, indicating the estimated time for design, certification and prototypes
• The budget estimation for industrial production
• A detailed and clear estimation of the final cost of the solution
• Supporting Documentation: Any additional supporting materials, diagrams, simulations, or research that could help in understanding and evaluating the proposed solution

Workshop

SAVE THE DATE! Join us for a workshop dedicated to this challenge on 12 February 2025 [DOWNLOAD THE WORKSHOP CALENDAR].

You will be able to listen to the Challenge Owner and the relevant Business lines involved as they talk about the challenge details, while the Enel team can answer your questions: don’t miss this opportunity to perfect your solution and meet the brief’s requirements.

What Happens Next?

After the Challenge deadline, the Challenge Owner will complete the review process and decide with regards to the Winning Solution(s). All participants that submit a proposal will be notified on the status of their submissions. However, no detailed evaluation of individual submissions will be provided.

The Challenge Owner will evaluate the proposal considering the Solution’s required features and characteristics, focusing the following criteria:

• The overall scientific and technical feasibility of the proposed solution;
• The economic potential of the concept (e.g., the Total Cost of Ownership);
• Its business potential for the Challenge Owner;
• Novelty and not obviousness;
• Potential for proprietary position (i.e., is the technology novel or protectable);
• The user’s capabilities and related experience;
• The realism of the proposed solution;
• The maturity level of the proposal.

If the reward includes the opportunity to collaborate with Enel, once one or more suitable solutions have been identified, Enel will reserve the right to start a collaboration, by way of example, on all or part of the following activities:

• Test execution;
• Supply of prototypes (if the solution includes equipment);
• Installation and site tests;
• Follow up and monitoring of the proposed idea’s behavior.

You will receive feedback at the end of the assessment.

If your proposal is accepted, an Enel contact person will get in touch with you to discuss the next steps.

The final reward for this Challenge is contingent upon the satisfactory completion of the pre-awarding process, including acceptance of the Challenge’s Regulations.

The pre-awarding process includes obtaining some documents from the participants such as the signed Challenge Regulations and the Counterparty Analysis Questionnaire (CAQ).