Machine vision is behind many of the great advances in the automation of the industry since it allows the control of quality of 100% of the production in processes with high cadences.
A non-automated process can be inspected by the operators themselves in the production process. However, in a highly automated process, inspecting the total production manually is a really costly process. Sampling inspection, i.e. determining the quality of a lot by analyzing a small portion of the production, has been used as a compromise solution, but due to the increasingly demanding quality demands of the final product, sampling inspection is not the solution.
It is in this context that the need to incorporate automatic systems for quality control arises, among which stands out the visual inspection through machine vision. The human ability to interpret images is very high, adapting easily to new situations. However, repetitive and monotonous tasks cause fatigue and therefore the performance and reliability of the operator’s inspection decline rapidly. One must also consider the inherent human subjectivity that makes two different people provide different results in the same situation. It is precisely these problems that can best address a machine, because it never tires, is fast and results are constant over time.
It is logical to think that the aim of a machine vision system is to emulate the virtues of people’s vision. For this, the first thing we must ask ourselves is, “what do we see with?” A simple question that common mortals would answer without hesitation “with the eyes”. However, the people who dedicate ourselves to machine vision would answer in a quite different way and say “with the brain”. Similarly, it can be thought that cameras are in charge of “seeing” in a machine vision system, when really that process is carried out by the image processing algorithms.
Obviously, in both cases it is a simplification of the problem, since the process of vision, natural or artificial, cannot be carried out without involving both eyes / cameras and brain / processing, without forgetting another key factor, illumination.
Many efforts have been made to try to emulate the human capacity to process images. This is why in the 1950s the term Artificial Intelligence (AI) was used to refer to the ability of a machine to display human intelligence. Among those capacities is that of interpreting images. Unfortunately, our knowledge about the functioning of the brain is still very limited, so the possibility of imitating such functioning is too. The development of this idea in the field of machine vision has been carried out by means of what is called Machine Learning (ML) popularized in recent years with the techniques of Deep Learning (DL) applied to the understanding of scenes. However, these techniques do not really have intelligence behind them, but rather are based on feeding them with a huge amount of images previously labeled by people. The processing that allows to classify the images as expected is considered like a black box and really, in most cases, we do not know why it works or not.
When machine vision is applied to the industry for the quality control there is usually not enough data to apply these techniques and it is required that the behavior of the system is always very predictable, so these techniques have not yet been popularized in the industry. That is why, when developing applications of machine vision for the industry, the objective is to solve well-defined problems in which cameras and lighting are selected to enhance the characteristics that are desired to be inspected in the image and subsequently endowed the system with the capacity of interpreting the acquired images with really low error levels.
Finally, the inspection results are stored and used in the production process, both to discard the units that do not meet the quality requirements before adding them a new value or to improve the manufacturing process and therefore reduce the production of defective units. This information is also used to ensure that the product met the quality conditions when it was delivered to the customer.
Among the different applications in which these techniques can be use are geometric inspection, surface finish inspection, the detection of imperfections in manufacturing, product classification, packaging control, color and texture analysis… and so on.
At CARTIF we have carried out numerous installations of machine vision systems such as cracking and pore detection in large steel stamped pieces for bodyworks, detecting the presence, type and correct placement of car seat parts, the detection and classification of surface defects in rolled steel, inspection of brake disks, detection of the position of elements for their depalletising, quality control of plastic parts or the inspection of the heat sealing of food packaging.
The quality control of the products we consume daily is carried out by means of reference methods that present great limitations as to the necessity of sampling (which may or may not be representative of the whole), which also entails the manipulation and even destruction of the sample (which is a significant economic expense) and does not offer us an immediate response, which makes difficult taking decisions.
The agro-food industry continuously seeks solutions of quick, simple and direct application in order to improve quality and security controls of food, both in the final product and in the different phases of its production chain, starting with the variability in raw materials.
The spectroscopic (near-infrared) NIR technique alone or combined with hyperspectral imaging methods and using chemometric tools in both cases is a technology that saves the tedious, costly and long laboratory tests that the product usually requires to control its production.
NIR technology in online mode allows the monitoring of a process and product without interfering in it, to carry out a continuous and individual control of the production and continuously supervising the quality of the product, which facilitates an immediate adjustment if it would be necessary, contributing directly to the profitability of the plant.
It is true that this technique requires prior preparation of equipment with a significant associated cost, but in the medium term, it is compensated by the ease and the speed in the response to this need.
There are a lot of applications in which agro-food industry has applied online NIR in its labs, for last 15 years, but very few have implemented it directly on the production line, where its advantages are clearly evident.
What would it bring us?
With the information obtained from each product in real time and on the processing line itself, we would facilitate the taking of decision to ensure its quality and safety.
Where do we start?
Identifying the moment in which the product requires the control of some critical parameter that ensure its quality.
How do we do it?
Creating calibrations for each parameter at each point in the process that we want to control.
In CARTIF, we are sure of these advantages because we have been working with this technology for more than 15 years. We have used it frequently to support companies in the agro-food sector, starting with a diagnosis of the process to identify in what way, how and when it is most advantageous and necessary its application and developing the methodology to implement it in the company.
During these years, we have developed a wide variety of applications for very different products: from cereals to pulses, feed, eggs, dairy products, meats, cured meat products, etc., saving important challenges in terms of heterogeneity of products and the determination of minority compounds.
Currently, in CARTIF, we carry on working to companies make the most of this technology and we go on developing new interesting applications for industry and, definitively, for the consumer, such as the identification of contamination of food with potentially dangerous products for sensitive people, whether due to allergies or intolerances.
The SMART term has become part of our life. Thus, if we introduce it in Google about 1.8 million entries appear, which gives us an idea of how widespread it is. Now, not only phones are smart, we also find this term applied to watches, televisions, homes, cars or cities.
It is an emerging concept and its meaning is subject to constant revision. For example, for new products that are released to the market, the word Smart is related to advanced technologies. So it is now possible to answer calls or check whatsapp in a smartwatch. However, in more global areas such as cities, the term “Smart City” is closely linked to sustainability. As Miguel Ángel García Fuentes comments in his recent blog, a smart city is sustainable and efficient in its ecosystem. CARTIF is promoting these processes of urban regeneration in 16 cities, through our R2CITIES, CITyFiED, REMOURBAN and mySMARTLife projects, which include interventions in the fields of energy, mobility or Information and Communication Technologies.
Hospitals are like small towns. As an example, a medium-sized health center such as the Hospital Universitario Río Hortega in Valladolid receives more than 250,000 consultations per year or 25,000 admissions. Hospitals are also large consumers of natural resources (water and energy) and large generators of waste. As illustrative data, a medium-sized hospital consumes per year as much electricity as the city of Soria, generates around 9.000 tons of CO2, the equivalent of 7.000 cars and if we talk about waste, the figures increase to 3 million kg per year. In this way, the health sector contributes significantly to climate change (another term we are increasingly familiar with).
During the last 2 years, CARTIF has been deploying this Smart concept in the healthcare sector through the SMART Hospital project, funded by the European Commission’s LIFE call. The document “Healthy Hospitals Healthy Planet Healthy People. Addressing climate change in health care settings” identifies the 7 key elements of a sustainable hospital: energy efficiency, green building design, alternative energy generation, transportation, food, waste and water. Among these elements, LIFE Smart Hospital project has selected Energy Efficiency, Water and Waste. Thus, the demonstrative experience that is being carried out at the Hospital Universitario Río Hortega includes the application of best practices and available technologies and customized training in each of these three axes.
In the energy axis the actions that we have already implemented include the optimization of boilers, air conditioning and ventilation of the operating rooms, or improvements in lighting. In the water axis, we have identified the streams that were being discharged to the public sewage system without being sufficiently contaminated and different measures for their reuse were proposed. In this way, reject from the water plant of the hemodialysis unit has been taken to hospital cisterns. In addition, the outlet water from the evaporative panels has been recirculated to the toilet flushing network. Just as in the two previous axes, the concept “Smart” has meant optimizing engines, valves or pumps, in the case of waste, the concept involves people. Thus, training has been given to the 2,500 hospital workers for the proper classification, segregation and collection of waste.
Throughout the current year, we will quantify the effectiveness of measures implemented, not only in terms of saved kWh, liters of water, kg of waste or euros, but also in the form of environmental indicators such as carbon footprint or water footprint. In addition, we will publish a “Manual on sustainability in hospitals” that includes all these actions and favors the replication of the Smart Hospital project to other hospitals, at national and international level.
It is a very promising initiative and is attracting a great interest among the different stakeholders involved. Thus in October 2015, the project received the second prize of the OMARS awards, as the second best action in environmental sustainability in Spanish hospitals.
From CARTIF we encourage other hospitals and large areas (airports, supermarkets, shopping centers, thematic parks, etc.) to apply this “Smart” concept, making a smart use of its resources and thus achieve technical, economic and environmental improvements for a more sustainable future.
Latin America and Caribbean (LAC) is the developing region with the highest urbanization rate in the world. Its urban population has grown from 41% in 1950 to 80% in 2010 while the economic activity is focused on urban centers (60% – 70% of regional GDP). However, despite their capacity to generate richness, almost 70% of people that lives in these cities are doing so in poverty conditions. Furthermore, if we also consider the environment impact of these cities at the same time of their high vulnerability to climate changes, natural disasters and financial constraints, we are forced to think about the sustainability of their urban development.
The theory about traditional development postulates that the industrialization triggers to a gap between urban and rural productivity, reflecting in addition salaries differences between the two areas, and promoting thus rural-urban migration. At the same time, this theory justifies that welfare indicators are better for urban residents than rural ones, because they have more coverage in public services and higher incomes. However, this theory is not useful to demonstrate the development pattern of LAC countries, more in fact when they have levels of urbanization substantially higher than other regions of the world. Urban population growth in LAC necessarily does not let to their inhabitants better living conditions.
Therefore the cities, even more LAC ones, are based on complex and interdependent systems that have defined a sustainability new concept. This new approach goes beyond environmental issues because include cultural, political, institutional, social and economic variables. Thus, it is necessary to develop methodologies that study cities as a holistic, complex and multisector system that will allow us a qualitative and quantitative understanding of the problems of urban development and management in the region.
Smart City concept is born from this challenge and we, in CARTIF, understand it as a new city model based on three basic concepts: life quality, sustainability and innovation. This city model use to involve information and communication technologies (ICTs), but mainly the definition of sustainable and cohesive territorial models with environmental, social, economic, territorial and administrative objectives. As a result, smart cities and resource efficient cities are achieved, diminishing costs and saving energy, improving the services provided and the quality of life, and decreasing the environmental footprint. The final objective of these smart cities is not to show off their advanced systems and innovations, instead of this they must provide to their citizens a better quality of life, and in a future, anticipate their needs solving any problem that could arise.
In this sense, CARTIF has been working for years to allow the transformation of “traditional cities” into “smart and sustainable cities” in Europe and, more recently in LAC.
Our model seeks an efficient and integral urban regeneration that achieves social, economic and environmental objectives coming from the specific priorities of each city, integrating innovative technological solutions in the different urban scenarios, with a large citizen engagement, stablishing the foundations of a business ecosystem to facilitate the deployment of pilot projects and their subsequent upscaling and replication.
We hope to see examples of this new model of city in many LAC cities in the following years. Meanwhile, CARTIF has involved the city of Medellín (Colombia) in a project funded by European Research and Innovation Program H2020, which seeks new strategies to renaturing cities through nature-based solutions. Thanks to this, Medellín will have the collaboration of experts to identify, in a first approach, the economic, social and regulatory barriers that impede this kind of integral projects in the city.
“It is April 21, 2011. SKYNET, the Superintelligence artificial system who became self-aware 2 days earlier has launched a nuclear attack on us humans. The April 19, SKYNET system, formed by millions of computer severs all across the world, initiated a geometric self-learning process. The new artificial intelligence concluded that all of humanity would attempt to destroy it and impede its capability to continue operating”
It seems the apocalyptic vision of Artificial Intelligence depicted in Terminatorscience fiction movies is still far from being a reality, yet. SKYNET, our nemesis in the films, was a collection of servers, drones, military satellites, war-machines, and Terminator robots to perform a relevant task: safeguarding the world.
Today’s post is focused on a different but relevant task: manufacturing the products of the future. In our previous posts, we reviewed the Industry 4.0 key ingredients, the so-called digital enablers. The last key ingredient, Cyber Physical Systems, can be seen as the “SKYNET” of manufacturing, and we defined it as a mixture of different technologies. Now it is time to be more specific.
The term “cyber-physical” itself is the compound name to designate of mixture of virtual and physical systems to perform a complex task. The rapid evolution of Information and Communication Technologies (ICT) is enabling the development of services no longer contained into the shells of the devices we buy. Take for example, digital personal assistants like Sirifrom Apple, Alexa from Amazon or Cortana from Microsoft. These systems provide us help with everyday tasks but are not mere programs inside our smartphones. They are a mixture of hardware devices (our phones and internet servers) that take signals (our voice) and communicates with software in the cloud that makes the appropriate processing and answers after some milliseconds with an appropriate and in-context answer. The algorithms integrated into the servers are able to process the speech using sophisticated machine learning algorithms and create the appropriate answer. The combination of user phones, tablets, Internet servers (physical side) and processing algorithms (cyber side) conform a CPS. It evolves and improves over time thanks the millions of requests and interactions (10 billion a week according Apple) between the users and intelligent algorithms. Other example of CPS can be found in the energy sector where the electrical network formed by smart meters, transformers, transmission lines, power stations and control centers conform the so called “Smart Grid”.
The same philosophy can be applied at industrial environments where IT technologies are deployed at different levels of complexity. The fast deployment of IoT solutions together with cloud computing solutions connected through Big Data Analytics open the door to the so-called Industrial analytics. Better than providing theoretical explanations, some examples of the CPS applications at manufacturing environment will be more illustrative:
CPS for OEM manufacturers where the key components (e.g. industrial robots) will be analyzed in real time measuring different internal signals. The advantages will be multiple. The OEM manufacturer will be able to analyze each robot usage and compare it with other robots in the same or different factories. They will be able to improve the next generation of robots or give advice for maintenance and upgrades (both hardware and software).
CPS for operators: a company providing subcontracted services (e.g. maintenance) will be able to gather information on-field through smart devices to optimize their operations like for example controlling spare parts stock in a centralized way instead of having to maintain multiple local stocks across different sites.
CPS for factories: gathering on-field information from manufacturing lines (e.g. time cycle) it is possible to build virtual models of the factories and create off-line simulations to aid in decision support (e.g. process optimization) or study the impact of changes in the production lines (e.g. building a new car model in the same line) before deciding new investments.
The combination of physical and virtual solutions open the door to limitless possibilities of factories’ optimization.
There is no doubt that cereal grain are the main source of the diet of consumers around the world. In fact, global cereal production in 2016 was 2,6 million tonnes (FAO data) and account for 30 to 70% of daily energy consumption (FAO data). Cereal intake should be 2-3 servings a day, and, according to the Mediterranean diet model, the consumption of bread and cereal derivates (pasta, rice and other cereals) should preferably be done as whole grain form.
Cereal grains are a great source of carbohydrates, protein, dietary fiber, vitamins (especially from B group) and minerals. In addition to the germ and the endosperm, whole grains, in contrast to the refined ones, contain the bran fraction which is eliminated mainly during the refining process. Whole grains are a great source of vitamins, minerals and phytochemicals, and there are numerous studies linking these properties to the prevention of chronic diseases.
This clear evidence of the importance of the consumption of whole grains has encouraged that many countries recommend their consumption and in some of them, like the United States, the campaign has arrived at great restaurants and schools where all type of cereals are used with the aim of modify the consumer perception of “wholegrain concept“.
In Spain, despite these recommendations, most cereal products are still made from refined flour. This happens, in part, because the food industry finds difficulties to adapt the recipes from whole grain ingredients because the incorporation of fiber generates some technological problems. On the other hand, there is a lack of demand on the part of the consumers with an educated palate to certain flavors and textures in the canons of the refined products.
Undoubtedly, there is a huge need of training consumers in the knowledge of which are the healthy options. However, there is an awakening of the industrial instinct to improve the nutritional profile of cereal based food products, through the moderation of the physiological response that they exert in the organism (eg, through reducing the glycemic index) and through the feasibility in the industrial application of whole grains, improving their incorporation into products and reducing the detrimental effects of sensory quality associated with the incorporation of fiber
A technological revolution come up to rich this challenges and to develop new cereal products that bring a clear benefit to health, such as high protein content, high fiber or whole grains, with new sensory experiences, cereals and less common flours (chia, quinoa, legumes) and of course, rich and appetizing. More whole grains in our kitchens, in our tuppers and in our appetizers.
Some of the improvements that are being made by the sector in recent years and that put technology at the service of cereals are:
Fiber-rich pre-fermented doughs, has been shown to be an advantage in the production of bread, which allows the avoidance of defects in bread volume or crumb texture that direct application of the fiber can cause. In addition, the pre-fermentation of the mass of whole grains or fiber-rich doughs gives breads with less impact on the glycemic index.
Grinding or milling systems that remove only the parts of the grain that deteriorate the technological quality of the cereals and improve the retained concentration of bioactive compounds. This means that it is possible to obtain whole flours with more nutritional quality but without the detriment of the quality in final cereal product that causes the incorporation of bran.
Softening processes of whole grains; they are processes prior to the incorporation of the grain in baking dough that allow the appearance of whole grains with a soft and pleasant texture inside the dough and that do not affect the baking process.
Another option is the of the fiber aqueous extraction, which allows the same nutritional and healthy benefits, without the detrimental effects of incorporation of fiber into bakery products.
The fractionation technologies allow the production of ingredients rich in β-glucans easily applicable thanks to better physical properties (like hydration or viscosity) and technological.
Another technological improvement to be able to offer all the value of the whole cereals is the transformation of the part of the insoluble fiber to soluble through the application of the extrusion technology.
There are many whole grains, not just wheat, waiting for their opportunity to apply the appropriate technology to favor their use while maintaining their nutritional properties and giving rise to new products based on whole grains.
