Software Technology

Hardware Technology

We have the resources, expertise, and state-of-the-art technology to process today’s large, structurally complex data volumes.

We are continually improving the capacity of our hardware systems to meet the needs of our clients. Our Linux-based processing centres in Calgary, Canada and Noida, India are configured with an architecture of HPC Clusters and easily expandable hardware. This design supports future industry standards and demands. Our systems are aimed at optimizing turn-around for computationally and data demanding applications.

The computer systems are complemented by high-capacity, high-throughput disk farms. Using innovative technology, these are networked together to provide users with a unified data space for efficient handling of large projects.

We proactively protect our systems and databases and follow our Company Data Protection and Cyber Security Policies to ensure the security and redundancy of all data that is entrusted to us.

An important element of our system design allows for secure remote access as an added convenience for our customers. A seamless extension of our in-house hardware is achieved via our relationships and experience with providers of Cloud computing resources.

Cloud Technology

Cloud services provide additional server and storage hardware on an as-needed basis, allowing us to deal with any volume of seismic data or time constraint that may arise. Absolute is an Amazon Web Services (AWS) Partner but our workflow is easily adaptable to any Cloud provider including Google Cloud and Microsoft Azure.

By accessing the Cloud, we can utilize state-of-the-art infrastructure that includes the latest CPU & GPU, Cluster, and SSD technology, allowing virtually unlimited resources on a project by project or even application (such as PSTM or PSDM) basis.

This allows our client to choose where their seismic data resides while being processed by Absolute; in Canada, or in any of the global regions where the Cloud provider has a datacenter. AWS, Azure and Google currently have locations in numerous countries worldwide. Cloud providers maintain the highest levels of infrastructure and data security, implementing the latest in technology and methodologies. Data integrity is maintained by redundancy across datacenters within the same region.

We have built Linux kernel images for utilization on cloud infrastructure with all the required software and libraries to support our proprietary and third-party applications for seismic processing and advanced imaging. This allows us to quickly start immediately usable production servers as needed. Security is ensured by utilizing a Private Virtual Network (PVN), maintained by firewalls on both ends.

This robust combination of efficient workflows and scalable hardware access allows Absolute to provide our clients with the timely and cost-effective data that they need in order to make accurate exploration and development decisions.

Machine Learning

Deep Learning is a subset of Artificial Intelligence (AI) approaches, inspired by the workings of the human brain, which aim to process data for a large variety of applications, ranging from detecting objects and recognizing speech to making decisions and translating languages.

With the increasing popularity of deep learning, the Convolutional Neural Network (CNN) has been extensively studied and shown to yield promising results in image analysis, as CNNs process data with a grid-based topology in order to keep spatial information. Unlike conventional neural network structures, where only one-dimensional data is accepted as input, the CNN is designed to process multidimensional data. It consists of an input layer, output layer and hidden layers. The hidden layers of the network consist of convolutional layer, pooling layers, fully connected layers and normalization layers. Using different functionalities of the layers, the CNN not only keeps the shape of input data, but also powerfully extracts the features that can present the image.

Absolute Imaging has developed technology for the application of machine-learning to the problem of velocity analysis. The process involves the compilation from historical projects of training sets consisting of semblance images together with the corresponding expert-picked velocity functions. Training semblances are separated into overlapping strips or patches and each patch is labelled with a velocity value obtained from the associated expert picks. These labelled patches are used to ‘train’, that is ‘iteratively to refine the parameters of’ a CNN model that can ultimately be used in predicting optimal NMO velocities for any semblance image given to it. Many such training patches are needed as a CNN typically involves dozens of layers and millions of initially unknown parameters. Individual network layers may consist of anything from classifiers to some form of image-processing filter. The proposed model uses the very successful VGG16 network and further improves it for the application.

With a well-established training set in place, the program can analyze new semblance images for fast and reliable velocity picking based on its past experience. Early results have shown promise for use of the technology in obtaining useable preliminary velocity fields. With continual addition to the training sets, we expect the power of the tool to increase both in reliability and scope of application.

In the future, the study can be extended by taking other machine learning schemes into account to improve other aspects of processing workflow. Remote sensing data such as seismic and well logs is usually analyzed visually by geophysicists in order to characterize geological and structural properties. This type of pattern recognition fits well into the Machine Learning (ML) and computer vision field. For example, deep learning can be used to discover and correlate stratigraphic units across multiple wells by recognizing patterns, textures, and similarities in the data. ML methods, specifically Support Vector Regression (SVR), can be developed for reconstructing seismic data from under-sampled or missing traces to facilitate 5D interpolation. Seismic noise attenuation is another common obstacle that can benefit from the strength of deep learning in adaptive denoising. Absolute Imaging is working on finding new ways of using data analysis and machine learning in current and future applications.

Technical Library