A picture is worth a thousand words.
Computer vision is computer based automatic interpretation of images. Machine vision usually refers to computer vision applied to an industrial application, but it can be used for many different types of applications across different business segments. I have experience on these:
Full stack of machine vision software
Are you wondering how you could use machine vision or artificial intelligence to support your business? Would you like to improve your product development processes and apply best practices? I have over 25 years of experience in the field, working in product development leadership roles across several companies and conducting research at a university. I can help you uncover the benefits of new technology without falling for the hype, and guide you in avoiding the pitfalls I've encountered myself — there have been many.
Do you need assistance with designing the architecture for a new product or product version? I have designed machine vision software and complete systems for multiple companies. My experience spans from embedded, heterogeneous computing platforms to cloud services and user interfaces. While I’m not a top expert in everything, I have a solid grasp of a wide range of technologies and a deep understanding of complex systems.
Do you want to find something in an image but don’t know how? I have decades of experience in software development, computer vision and artificial intelligence. I can develop precise measurement software, AI-driven analysis algorithms and intuitive user interfaces for these. I ensure that the code quality and performance the code I write are top-notch.
Flexible service according to your needs
I provide both packaged training courses and trainings tailored to the client’s needs at a fixed, pre-agreed price. I can guide both management and product development or project teams in the intricacies of machine vision, data processing (including GDPR), software development and AI methods. The training format can range from lecture-style information sharing to hands-on coding exercises, but it always remains grounded and practical. If you wish to get empowered by AI hype, you'll need to look elsewhere.
I primarily conduct consulting and design work on an hourly basis. The deliverable is usually a well-written, illustrated document in either Finnish or English. Design work may also include software demos. If needed, I can also develop complete software or components on an hourly basis. I invoice my work in agreed-upon intervals and provide the client with a detailed report of my activities.
Sometimes pure subcontracting is too expensive or risky a solution. On the other hand, I might already have software that fits the application. In such cases, I can sell the software I’ve developed under a licensing agreement. The license fee can be a one-time payment, per installation, or per user, depending on the situation, and it may include the software's source code.
An effective way to elevate your skills to the next level
What exactly is artificial intelligence, and what is it not? Learn to understand how AI works, where it is applicable, and what its practical limitations are. What should be considered when handling data? This course covers AI in general but focuses on computer vision. Alongside general knowledge, practical experiences from the field are discussed. Through exercises, you will identify areas in your business where the technology could provide measurable benefits. The course is designed for a general audience and does not require technical expertise.
Get familiar with the basic concepts and methods of machine learning. The content emphasizes artificial neural networks, which are are the backbone of most modern machine learning models. The course does not delve deeply into theory and algorithms. Instead, it aims to create a practical, intuitive understanding of the technical aspects of machine learning. Strong mathematical skills are helpful for grasping the material.
Learn in practice how to create a deep learning neural network and how to use it for visual image classification. You'll learn how to train machine learning models and use them to classify images into predefined categories. Knowledge of the Python programming language and machine learning fundamentals are required. If these skills are lacking, the Fundamentals of machine learning course is recommended as a foundation.
This course provides the basics for deploying machine learning models in production on the cloud, PC, and embedded devices. You will learn how to optimize a model to ensure sufficient computational power and how to prevent unauthorized copying of models. The course also covers best practices for data management and model development. Coding skills and knowledge of software development fundamentals, such as version control, are required.
Removing optical distortions is essential for accurate measurements. This course covers the basics of image formation and optics and provides practical training on calibrating different types of cameras. You will also learn how to combine multiple cameras, for example, for 3D measurements. Programming skills and the ability to operate a machine vision camera are required. Proficiency in matrix calculations is beneficial.
Can't find a ready-made training package that suits your company's needs? No worries, training can always be tailored specifically for you. I am familiar with software development more broadly and happy to share my experiences.
My name is Topi Mäenpää. I founded Konekatse because I believed I could best utilize my full expertise by offering it to companies facing challenges in machine vision and image analysis. For you, it means the opportunity to quickly elevate your company's expertise to a new level, offer better products to your customers, and get ahead of your competitors.
Three reasons I believe I can help:
I have a Master's degree in Computer Science. I pursued my doctoral studies in intelligent systems and defended my thesis on machine vision in 2003. I have a solid foundation in the mathematics required for the field and have written AI algorithms from scratch when ready-made libraries were not yet available. I have authored over 20 academic publications and have worked in research roles as well as assisting in teaching AI and digital image processing courses.
I founded my first software company in 1997. I have been developing machine vision software since 1999. I have been involved in four start-up companies, achieving success sometimes but often facing failures. Working in small firms, I have seen the entire spectrum of business operations and product development from customer interface to the lowest hardware-facing bits. I have built product development organizations and processes, defined and designed software and systems, and coded all their components myself. I have contributed to and deployed several machine learning-based products.
I began exploring computers and software as a young teenager, and my enthusiasm has only grown. Alongside my work, whenever time allows, I have coded and assembled various devices. Over time, I have gained experience ranging from small embedded devices to cloud services. In addition to optimized algorithms, hardware interfaces, distributed computing and backend services, I have developed desktop and web user interfaces as well as games. I have written production code in multiple programming languages for different operating systems.
Please call, send an e-mail or fill in the form below.
Konekatse Oy
Leppiniementie 12
13500 Hämeenlinna
Finland
First name dot last name without umlauts @konekatse.com