CORDIS - Forschungsergebnisse der EU

Intelligent System for optimising the on-line Finish process for stone slabs and wood panels (WOODSTONE)

Final Report Summary - WOODSTONE (Intelligent system for optimising the on-line finish process for stone slabs and wood panels)

The general objectives of the WOODSTONE project were:
- design, development and testing of a system for acquiring information on stone slabs and wood panels during final manufacturing phases and for assessing final product quality;
- the system will be made up of a digital camera, 3D laser scanner and hyperspectral device; information acquired will be collected and elaborated by three specific SW systems.

More specifically, the technical project objectives consisted in:
- supply reference parameters to final material manufacturing phases that influence finished product quality level: semi-finished products are examined and the machines are informed on which operative parameters to respect on the basis of the desired qualitative standards;
- supply objective data to the finished product classification phase divided according to various quality levels: finished product quality may depend on colour considerations and also on number and type of still existing defects on the finished product.

At the moment, both the stone and wood industrial cycles are composed of various operative phases. In both cases, the industries produce semi-finished products that need to undergo final finishing phases: polishing and planning. These final operative phases, according to the working methods adopted, may influence the finished product quality level. In the greater part of cases, finishing is carried out in a standard way or is entrusted to experts' common sense. The project has been conceived by considering this situation. From a scientific point of view the core of the issue consisted in knowing how to adapt modern technologies (three-dimensional (3D) scanner, hyperspectral analysis and digital camera) to the project's specific needs.

Two main problems needed to be tackled:
- data acquisition phase: the equipment used must be interfaced and adapted to a particular type of data acquisition phase. The material defects are quite different, the material dimensions to be assessed (precisely) are large, colours could be extremely important, etc. In other words, assembling and adapting the individuated components to the project aims is in no way commonplace.
- data collection and elaboration: the interpretation and the need to manage and link various types of data in such a particular field, is very demanding. The dedicated SW systems must characterise all the various types of existing defects: this may be carried out in strict connection with the end-users' experience; the 'significant parameters' that must be determined together with the end users must, in view of the subsequent finishing operations, be able to completely characterise the acquired defects.

The overall project was divided into the following work packages (WPs):
WP1 - Characterisation of the state of the art (future end-users belonging to the stone and wood sectors; advanced technologies available on the market that could be of interest to the project prototype and its components);
WP2 - Characterisation of technical specifications;
WP3 - Prototypal system design;
WP4 - Prototypal system implementation;
WP5 - Testing and optimisation;
WP6 - Assessment of results and exploitation plan;
WP7 - Dissemination;
WP8 - Management;

The system, made up of three different types of equipment, was mounted upon a frame of cast iron pipes. A specific software system was set up for each part of the equipment, while the developed lighting system was suitable for the overall system.
The WOODSTONE prototypal system was implemented in the industrial premises of the Italian small and medium-sized enterprise (SME), project partner, Cogemar and is made up of the following components:
- a metal structure with Innocenti tubes with maximum height from ground level equal to 4 metres;
- The WOODSTONE metal structure is positioned at the end of the traditional polishing work cycle, on the roller conveyor normally located at 80 centimetres from ground level where the stone slab or wood panel is placed for the polishing phases;
- eight 135 cm-long and 80 Watt neons (four on each side) located upon 2 Innocenti tubes positioned on the outer side of the roller conveyor directed towards the opposite-facing structure sheet; specific deflectors are used for diffusing light and avoiding reflections on the stone lab/wood panel;
- laser scanner, digital camera and hyperspectral equipment, positioned at the structure top upon a connecting Innocenti tubes and specific supports;
- devices for modifying the positions of scanner, camera and hyperspectral equipment according to the slab/panel to be acquired; the scanner, for example, must be positioned at 2.50 metres from the slab/panel; this distance must be modified for the other components;
- white sheets positioned on the four sides of the WOODSTONE structure to enable optimum image capturing of the slab/panel;
- SW interface equipment capable of setting up a quality control methodology for final quality assessment and automatic standards classification;
- the prototypal system assembled as described above and with the fully tested system components will allow operators to carry out all acquisition and data elaboration operations in no more than 20 seconds, considering around 10 seconds for acquisition phases and 5 seconds for SW data elaboration operations.

The test results showed that from an industrial viewpoint, the most effective and promising results were obtained with the digital camera and, consequently, with the specifically developed software. The camera, in fact, is able to give overall qualitative and quantitative information on the interested wood panel or marble slab: this is important for firms and companies working in the wood and stone manufacturing sectors because interesting classification information and items may be acquired which is important from a commercial point of view. The laser scanner and hyperspectral system, instead, although able to analyse a single defect and give detailed qualitative information on such defect, are not able to give the same general commercial information as in the case of the digital camera.

The 3D laser scanner enables to study and to measure area and distances in deep detail; the depth of a defect can also be thoroughly evaluated. This is interesting for analysing defects and for creating a database when handling and treating valuable materials. The scanner however also presents flaws: firstly the sample needs to be still when acquiring the slab/panel image. Although acquisition is very rapid, the working cycle is interrupted: this is not particularly convenient from an economic point of view for low-cost materials. Furthermore, the image acquisition does not cover the entire slab or panel and a series of acquisitions are needed to acquire the whole image. As mentioned above, this could be conducted for particular and specific cases (the need to study a particular defect, valuable materials, etc.) but not for everyday industrial work.

Therefore, although the laser scanner laboratory tests achieved excellent results, it presented the two foregoing industrial problems:
-the hyperspectral system gives excellent and thorough information on the materials composing the panel or slab. The system, therefore, is able to give details on the internal parts of the panel/slab and on the composition of a defect. From an industrial point of view, however, a main defect was individuated: the application field is quite small. Again, as in the case of the laser scanner, this system is interesting in order to acquire specific information on particular materials or defects.
-the software set up for the hyperspectral system gives information on material/defect composition through a series of graphs. Regarding system positioning, given its reduced size the hyperspectral system was positioned next to the digital camera.

The development of this innovative system will allow the end user (stone and wood manufacturing industries) to acquire a competitive advantage as to maintain or consolidate the leadership that EC countries currently own respect to the non-European competitors. The system manufacturer SMEs will obviously open their market position as suppliers of inspection equipment in the stone and/or wood machining sectors. This will be obtained thanks to the envisaged engineering and industrialisation of the innovative WOODSTONE system characterised by performances that cannot be actually found in the stone and wood industries. The possibility, in fact, to develop a system, and the related analytical processing strategies, able to detect, directly during the finishing processes, the characteristics of the surfaces both in terms of their aesthetic attributes (i.e. surface patterns, roughness) and/or their reflectance properties will permit to control the process on the basis of pre-assigned rules, optimising in this way the entire process without any human intervention.

Such an approach will allow the manufacturing SMEs to produce more standardised and homogeneous products, envisaging the possibility to recognise as 'objective (quantifiable) quality certification' a material attribute, i.e. the surface finishing status, strongly influencing the quality of the final product, so far completely ignored by producers and by the market.