and the ladies and gentleman and it's my blessed to be here in sandy uh

but no and in this uh conference and then i uh i have a real

my presentation uh

i had quite practical and at the channel level i'm an engineer i'm not the

mathematician so hopefully this is okay to you

so um

and i will talk about the mess envision solutions and especially if although forest the

industry

and uh maybe the next slide that was more about motivation this is accomplished every

of my university in that in that a lot better and the and this area

is that so called in is like this trick area so it's full of the

lakes and forests

and the forests it is a big business big in the study in that area

so actually it's the area where there is the most then set at a number

of while males and solving this in the world

so it's quite natural uh since the six kilometres from our university there is a

big by a male and there are two other paper mills quite close

um about um

L one university uh it's one of the three a technical universities in finland and

here are our main focus areas

so green energy and technology

sustainable value creation and international have of R S and relations and you can see

because we are at the border is the closest to university to rest of the

so that's quite a match

so um and also my uh research area in must envision connected the forest industry

it's quite close to these calls

as you can later like to see

so this is our motivation this is probably very clear to you um what is

the motivation of computer vision and machine vision but from the practical point of view

at six i especially these and these are not of fake pictures these are real

pictures this is a person in uh mail paper mills laboratory to in serious uh

quality assurance testing

so it's mostly damman wiley and you as you can see it really minor

so there are some are based but it's

it's uh bring it on a piece of paper board and then this person is

used in magnifying glass checking

is it is it okay is the quality okay what this the prediction of the

usability of this um at the

and that this is how it should be done

so bill built in uh matching wits that's the same thing and the in the

standard last way one of the channel and C is there is that this depends

very much on human body and the uh and uh be right is based on

individuals a preference is still there is a slight variation

they are and also this monday morning and uh friday afternoon effect too

and this is a standardised way to do it

since this information goes from paper mail yeah males from this uh paper manufacturers to

a printing house that what is the quality of the product is any

and actually printing houses are doing their own similar tests so as to where E

Y

that opinion these the same

so this uh would be a standardised solution is

so uh

another thing of course is that this is similar don't knows task and some that

cell or spencer's at in the industrial environment

and uh this motivations i think you have a scene that it's fast the more

accurate less expensive i wouldn't say gee but i would say less expensive because that

must be sent to quality and easy to use is important

this

um people who will be systems they sometimes will get that it must be easy

to use

and sometimes the realm us to do not some in for my mation or two

months options which you never actually

so is it the use

and it's a good thing

and motley question of course is the replace human recent totally in this case that

somebody ways

so no human of we need it anymore at all a mess in instead

ses human-based and so sometimes to keep some hint this is mainly for example in

medical image processing

and then perform that's the time possible to human vision system you can imagine some

online uh solutions where you simply don't have ability to be so fast

when a paperweight is running the that the meters per second

so you that you can't handle that time sequence in your mind

or simply we don't have always in system can see spectral information for some

so this is our um

uh

some kind of uh starting point

and uh

this is just same shortly about the low-level auditory machine vision and but the recognition

of auditory we have we have around the when the people

doing this results and we have a it you kate the twenty seven uh doctors

in this area so in machine vision and many times the uh these theses they

are connected to real problems it's comes from industry go from bases

and we have around one million euros but here is how much it and six

hundred thousand euros just

is um

um external funding and of course L one bit main goal is to produce a

speed of common one is where these ideas could be implemented as soon as possible

in practice

um generally however is not only supporting building a built in a visual inspection systems

like here we have a many other areas like a fire model correlation here is

a bacterial a matter

so that sensors have a it's a consist of that there is not on any

electronics and then intelligent robotics and object recognition here face recognition and one area which

we especialised this medical images analyses and especially analyze civility not images so to see

some uh this easy some problems for example cost by diabetes

so that is also area

um if we think about the visual inspection actually it's uh um

it's a it's a actually what white lots color correction so there are many other

universities involved and also many real eight comp running so a lot of a lot

of all operation is needed and if we need to think about how one of

our main objectives is the develop novel results efficient and environmentally sound image in image

processing and image and the lies in met thoughts for machine vision applications especially for

overall quality assessment and control so the idea is to all these single measurements together

to be able to control in general

the whole industrial process

and we aim to offline and online solutions obviously that is a lot part of

our example in the preview on the previous slide that it is offline

uh solution because it happens in the laboratory but then we need also online solutions

directly uh in the industry

and mainly we are focused on forest and printing industry

so at the lower product already and the factory

about thirty level

and um if we think uh how

what is actually happening there you see that this contains look over different the industrial

is text there so it's actually quite complex uh combination before you end up to

get to have a lot of paper there are so many things happened before that

so in bright this we have a some would might be real then we have

well and then paper and finally some print so something you can print or

uh and uh if you think wow why to do it okay the results efficient

uh resorts in efficient uh environmentally sound production with no one data using less rollment

the real water and energy and actually this is a cool thing because this to

be environment three uh one shows uh use the same as the road using it

efficiently so there is no contradiction between these things because if you use less raw

material water and energy you save money and you also save environment

and the no other known quality means that the uh you use should put quality

as low as possible and it still feasible quality so it's still suitable quality you

can always book the parameters so high that the quality is high but you waste

a raw material what the energy and so on and you waste money

so as low as possible but still reasonable quality that's the idea so that these

than on quality and we'd without having some measurements and they interpretations we'd be gone

we don't know what these what these that quality so that these what we try

to write the so

actually our uh meals in finland they uh environmentally in very good condition because the

lakes around a white lee

and that we actually condo for two only you of lakes because then the for

its we like to do so we must really think of environmentally seems to

so um

idea simple in the way imagine they the automatic analysis and the in the forest

industry

uh we talk about this will also line quality has assessment business but this done

right traditionally manually

and the campy time calm you consuming and labourers to

then we have online quality assessment and then tool for real time control actually these

two areas and the research done their support this idea

to have a some kind of real two-dimensional uh quality control based and to advise

dimensional units

and uh what we have a

study we have had we see this as a as a collection of uh research

approach it so we invested

so we have one uh rotate which is dealing with power

so how to put this uh raw material in the bases for paypal make

and what could be done there

then the one thing is that what can we do um

connected to a paper messy so when real paper making is happening so and there

is a long paper my signal for one hundred be there with the long volume

along more what we can do there

and of course these are connected so if you know about the quality of raw

material here you can you can put this information here and you can take advantage

of that

and then this is your final product okay um here you can't do anything anymore

it's already they are so paper so after that you can only predict how to

print on and what is the pretty that uh applicability and run ability when doing

some pretty

and then

then uh we test the quality of the paper and actually finally interesting thing is

that how we see that you meets or something pretty become paper that can be

mortal human beings set preferences is connected to some physical much innovation measurements

so that's is that and the system will be nice that you put the print

into that find a much emission system any else that how human beings are uh

but if you're in this but it's

"'cause" uh the quality of an image but in it of course it's very important

you don't want to see some bad quality images but what is the bad quality

and what is that what did when i ask if i thought make the test

we you i assume there are some but em sis but there are some form

of that are still so this is a big question and this actually relates to

any image information problem whether it's print date or whether it's speech it's exactly the

same problem

problem

it's very uh and fess in a be much fascinating problem actually and what typical

i will solve some results about this before and the steps here

and what one wants to remember this that uh need for these kind of research

must come from the industry so every single project we have a start date has

started in that way that we have once industry what do you need and then

we have stopped at the check what we can

so it's actually it's not uh selling your ideas it's uh

asking needs and uh formulating the questions

okay us then i would like to talk about some um

things uh which we must consider and one is to recognize imposes phenomena

so actually what we should and this that is the question and the who knows

the ground rules and why and how

and

yeah and it is it really possible the model the expert knowledge

and uh this is a good question because we must we must talk to be

a those experts there is that the what they would like to know and what

we could see and uh

sometimes it's islands in if the data that is very mobile so even the expert

have an ever seen they the from some certain industrial processes

so this really needs need something in and actually i have an all this when

we are having them some uh comparisons that the and i is that where is

the ground rules if i'd use your data and uh and some might be just

looks good but that's not you know

you should really be fine ground and maybe to other researchers publish your mate complex

operation ground

of course in this kind of uh industrial pros project said this comes a little

bit be late because firstly analyze the possibilities to make a real part that are

based on that it is it's but it will come anyway

and is this possible that's and that's and other is you sometimes it might be

very seen that the sum intensity values that everything but

you don't know before you think it's

mister in optimality

so what utterances in image and so uh there might be some illumination problems or

it might be a typical place in there or you need some uh not reflection

information but that some transparent information

and then we call to this uh do we need multimodal information that many times

we need so we have a transmission in its we have reflection immediately may even

have some extreme images

we call the different scales there and how to put this together so we have

multimodal information and how to how to put it together and whether it supports to

each other

then understand in measurements

so a bible college on unclear categorization some more clustering need it so that we

know enough about they may not then we have the first define what are the

classes there

so clear when it's be about the like categorization then it's just classification but many

times as a question of this especially when it's number of data that what are

the other clusters

so this might be a challenge to all of course accuracy and computation time if

you think of this uh paper with running that the meters per second you have

two options you where infrequently uh to sampling but the image resolution is uh low

or you have a high resolution but then you don't do it so frequently

and our idea is to go to this overall quality management and approaches process control

so step by step we more accurately know which kind of uh paragraph we will

get

and this is this is about how been and actually this goes into those of

previous steps as

as i mentioned

and there are some certain uh process phenomenon so probably means that the uh we

make a raw material for people making

so we need to check some certain things like uh i other some extra particles

they are which are not wanting to be there and the basis actually about suspense

and he meets so it's a liquid these is the right this is the right

about C have been some i'm one D the objects

and uh number of a number of this and what they do or distribution of

this on one be the object of like say that well

so mess are in these things

different kind of uh i things even for a sense and spectral image and then

this analysing part

so here are some examples here and one thing you know very important thing of

course is to understand the uh fibres that fibres uh distribution and position

so i'm not will industrial data for the uh

by new image and solutions and uh i'll idea is to pull inside the pipelines

which is uh have not been done so our data is totally novel so wipers

are you know interest the particles buckles and sales these bubbles the lapel approaches there

is some you know their problems there

and then how the form and the model the ground rules and this need something

can because if it's a mobile make the so there is no a human expert

that's believable expect is about

but the there is six but is about the uh i mean

so that can be used

uh and uh then uh one thing there is a segmentation the fibres another particles

in suspense and emits is still this is really weak

lee we thing and we have a done that the uh results and why this

piper's must be uh and the last is that you should know the uh three

species there it may sound strange that don't they know what is there but it's

not necessarily always the same tree and same kind of three material so uh the

know about these three C Cs is important and again this means that the this

viper is broken in the back leeway that later when you breed

how can they need that kind of uh fibres it's bad property

so the to detect number of things

is important

and we have some applications about this

if you call to my list of uh applications you see that should have that

six

then another use you uh not only five course but problem side important because particles

tell about some approach the situation and how they daily the daily and in that

way that um when there is a certain amount of purple bubbles the process is

uh okay so that such a certain size and certain volume distribution and how they

do it now when they are aligned in a factory they do it in that

way that they'll and the process for let's say two hours to be sure that

everything is okay but one hour would be enough

but you don't know that if you don't have an image

nonetheless

so you put even got terms is processed by means of halls and that would

be you say in

but you need in leeds analyze the solution for that so we have real the

field of that kind of a solution and we will shortly present this is very

impressive results and icp our conference in japan in november this is a couple operation

results with the tech technical university

we propose that matters

and actually this is a this is a quite challenging problem although that looks so

originally quite easy and if you have a some uh proposals or something to the

to this problem i'm very happy here you or your I S

so it is not easy

since it because these problems are not that we're a circular problems so they may

be deformed so actually we are dealing with the P Cs or for some kind

of circular say and we need quite the white mats thinking how to make it

that efficiently and then we must detect the problems in order to be able to

estimate what so from problems we estimate the volume has some value there and sof

detection of problems as important than in this case of course detection of the become

problems is more important than the than the small ones

but also this distribution map cell from distribution experts can tell something

and uh we have actually ended up matt's of thing you know feel that in

how we get some kind of suitable it speaks

and enough of those and how we start to combine of these pixels because then

never ideal unfortunately it's a liquid and it's also problem of three-dimensional imaging

there so but uh i must say that then we have got quite a very

good results and surprisingly this uh this uh false negatives some false positives they start

to compensate each other and uh not most of that we can detect right so

we can really get true positives but the so surprisingly um well then of course

the question these that what is the what is the ground and that because of

bases you we have given the experts to select a level of confidence

the whether there is a war or and solar or totally unsure

because it's not so clear what his answer to so

so these contains all the nearly all the possible problems i enumerate the in the

couple of slides ago

okay but that the that the say that the pc is challenging from the uh

point of view we mix process

some people sometimes say that all you test an application itself see but we are

dealing with the fundamental issues of image processing

okay then and that other thing that the uh okay fibres

um couples

so five resigned the liquid

and this is bob suspension bubbles in about suspension but this is the right biopsy

so when the power was ready each tried to be more to uh to a

paper mail and this is really a tried C and E there should be only

uh suitable rollment the we only but all these so with this dog you the

there's something and one it's um the particles because of what's something's there might be

even this and they are

so any anything

uh

the right of course the different kind of categorization even plastic

or some unwanted what would my

so it would be good to categorise these

different kind of uh the particles "'cause" that's there was the quality of the biopsy

if there is no the particles it's a it's a great but the problem she

for paper make

and so we need to a here the segmentation and then classification of these this

that the particle physics

once again we are dealing with the quantum fundamental issues of image processing and analysis

and here is a slight problem because we must do this testing that the laboratory

label

see scenes uh

uh if we do it online

then uh we wouldn't have time to do analyses so what we have done we

have a uh at the laboratory we have a creative israel about six and in

that way that the experts have really put the ground rules there

and then we can test whether we can recognise this separate that the particles

at the moment there is no analyses

you just put up sixteen and you hope that you will get would like

so that's motivation to build the system to do that and S

okay then let's go to the know now we worry at in the powerpoint so

a would deinterleaving liquid and do some processing there it can be column mechanical then

you get that right biopsies then you put these power seats uh to a paper

machine and then you start to make paper

and the um main question is that it is that it presents and this represents

a paper a bit so that what is running from the beginning to this and

when the product is ready uh the question is that can we see here or

something

something which we should uh is that the react to something and actually um now

there are some line scan or switch to body weekly scanning but that it's of

course different than the two-dimensional to time instant no uh real time image and uh

because of scanning

we actually can on see this kind of till the problems which you may see

basically simple maybe you somehow can see that and uh of course of human being

can see that's that the meters per second so close to plastic on C so

you're we must rely on these mis ordering systems and line scan risk on to

it because the coast line by line so idea is to take a real the

dual diamond two dimensional units uh challenge there is that we need eight income it

is

and put all these images together

and so it's quite the real-time problem and also build in an image problem too

because of uh some uh geometric uh imaging problems and things like that but that

we can do so we can find

fine these kind of because these problems these sideways problems they are because of some

uh not so accurate tuning at the beginning of the process and you can tune

the parameters there but you must know in which way so this gives feedback

they're to do this do you

so how to produce and use real-time two dimensional web white image state the problem

of or on online uh control

um then we can go uh further in our analysis uh we can we can

focus on at some uh more accurate level uh and uh get some information and

why we uh this is some kind of closer close up a information from that

with uh paper web that area

and there are some uh phenomena we should see that how actually the white borders

are grouped there that's important but if that my signal then we can then we

will get nice paper and print on it then we apply but there might be

some grouping problems and this that was also about

do

and uh here we can do this a multimodal problem because we have our lower

low resolution high resolution images we have a reflectance it means it means we have

transmit dance emits we have seven of images and how to help able to put

these things together and we can have some other kind of measurements to so we

have a bit different kind of a units in many thoughts and how we can

put them together so here for example the segmentation and analyze of three five or

flocks using transmis tense images

and when the orientations are nice then it's okay maybe when you have a at

the right that the other paper you know this that it that the there is

very nicely there some certain direction about another direction that some and the that's because

of

on purpose it's made in that way

and uh

so detection of surveys orientation is so one thing to uh check and then of

course the challenges here that how we get these correspondences when we have a different

kind of image state that from the same

so we have a correspondence problem they are

so how to do image registration

and why we are interested in surface orientation is that it's affects directly but

and then after we have got this paper then we can test what we call

and we have build the system to do this that and the vices

and so is to do this

the similar kind of solution to everybody

not depending on the human brain

and let's see what we can see from that so we can we can have

a different find of uh measuring devices here and the what we i'm gonna speech

in this uh for one is an to be in and evilness of print in

misery you know name it so it's modeling so we should pretty that the uh

if we bring the blue or lost by their how well how you nibble form

this

and can have some kind of my sermon some estimate and it's called model in

the in the printing industry

"'cause" if that was crying is not uniform it disturbs you i and you think

it's about three i think this is maybe

but pretty but you can see if you if it wasn't

so that's a one thing somewhat yeah modeling detect the modeling

another thing is that the

how will this step putting in happens on the survey so it's called haley a

test so detection of missing dots in that this but it is bought to say

about this that this is mostly about some uh frequency feel that in of different

kind of a yeah yes so that's a technique and then the question is then

how we weight of those different frequencies

it's as the linear combination or are there some coefficients with the with the combined

or something like that here this is very simple test the new print of on

a paper sleep uh don't red dots and then you simply calculate which starts are

missing because it means that the so face is not so cool and this simple

man in this woman in the picture was really count in and where is that

when the at the missing dot from the beginning and he may shirt she mesh

shirt that in millimetres and role there that ninety seven millimetres from the peak that's

the quality while the longer the better

so and we also made it is uh there's the one we used it is

there's another and we automated the this column

so ago that was automatically this of course the question there's a one challenge that

what is the mission of how much must be missing or and then person be

left there "'cause" of course this is natural situation this is not by nist

then uh um some small be there is how they sexy carissa must of these

is about that how these uh

this actual but it looks

what we can see from that from a single small not

it's obvious where about the predictability this is a part of an ability in printing

the it's called the peeking test that you press harder and harder that the that

uh test sleep until it breaks

or some i'm a other that meets this will start to happen to

so for usually uh and this is the border it's they're starting to see some

phenomenon the subways and when it's a paper it's imprecise the der of

so it totally breaks and the question is that how long you grow until it

starts to add to your break and once again people are doing by and calculating

how many millimetres from the beginning to that place and of course it's a at

the request and we have this happens this a one D uh phenomenon of the

early

and we have a thesis about

this topic to so this terms of butter unappealing the longer the better

from the printed point of view

and uh here is an example about the speaking tests he actually here you see

what happens the paper it simply start to tear off and this is not what

if the value and what and or it usually neighbour will S is that the

overall but there will become sums of place the effects there and this is able

to a little bit harder to define because there is no except order here is

quite clear that simply breaks here and that's the distance

and actually this company's we sell in this device it's called laplacian technologies our doctors

have a used up list a all pole one and they yeah selling this product

uh for this text

and this is about this missing don't the problem that

this print starts here and you put in the pixels and then you start to

check that how many pixels up the uh missing and that when the L one

is the

is the definition i is that where these come from and i think in a

laboratory level in those paper means they have just decide that it's twenty

so it's some kind of practical

practical value but you should always have some but i think the uh this twenties

in that way that you would never go to the end

it's comes from that it's better not to say that like one of our doctoral

candidate set in the doctoral dissertation examination that why is this threshold speech P and

he replied that that's a good number

but i don't recommend this was there will prefer that with questions after that

okay then uh and next issue when it's really pretty print it on um how

we can uh estimate the quality so which one which looks better and why because

you may have a different print the brain there's different papers different images how do

you can say that which one is better

so what we need to we uh selected some S in recent measurements

and we have the department of psychology from helsinki university to do in experiments with

humans asking their opinions and then you know how to do it we don't know

as engenders but they know how to do it so they collect be it's a

human being subjective but uh

what if the attributes and then to be put them together to get some kind

of well overall image quality index and with this index we can we can have

that

double for example then images and we completely in order and hope that it goes

in the same way as human being would have prefect

this is quite challenging results research to

uh in this results we have a couple of things to do is of course

because we need to expect around rule so we need to digitise image so we

must the register very uh accurately this data test images units and that's a one

uh thing is this uh and uh occur missile inequality and uh locate in those

areas which we have miss that one doctoral thesis is about that

so the predicting human humans experience of image quality for print the image

and the then we calculate the this uh like a bayesian network model that

weights the all these simple things uh seem not seem well in a way simple

measurements affect the most and um it's maybe not surprising that close is one of

those modeling is very important and colour but this time it is a really experimentally

proven because we really fast

is a human being set a group to tell they opinions and these opinions were

really model and compared to

real but uh physical my matching pursuit measurements

so this is not the case this is really the result

and in this thesis by almost and allow you can be more

um then uh i'm approaching the and only a couple of slides left here an

actual level with you with numbers because these are selected slides not fifty so far

so um and that next question is that what i've mentioned is at that the

how well does this depend on which kind of image you

and yes it does very much

so then we would need the system to finalise what is the content of an

image

to have a

E meets content based this one lady

so it would change based on image content because many print this or see it

on screen it's it has different preferences we deal than scene this and this close

to very current hot topic uh visual object categorization so how to and especially in

this unsupervised manner because we are not interested what are the objects at all we

are just interested what objects are important

and uh

we don't even without knowing what though they are

well let them to affect this decision making

so we are dealing with unsupervised the visual object that the crisis and then we

need the ground rules so we did i tracking

and to see what are the salient features for human brains beings and one in

a thing is that what are these images how what is the perfect test images

that that's a that's a big question because uh you may have a some kind

of side effects if you don't the

landay's in which is carefully enough and that's why we have from all the university's

media technology specialists who were preparing these images for us

and uh

uh

and then uh we ended up really the to do this uh visual object categorization

results and um well idea is uh use in self organizing map uh of for

a clustering of these objects and finally get these categories so but the visual codebook

use in

self organizing map the component it and the also some uh evaluation techniques then we

have a new uh we use uh published new images uh it said so randomized

object so we make the tricks with the for foreground some background images and also

saliency detection

and this was connected to the sidetracking the results

and all also

also better techniques for what this categorization

and uh we have also published the uh totally new limits that it's dealing with

up static images where we are interested this ci tracking when we have a collapsed

of images so we don't have any background to

there we have learned to do something

and then we take that how i look at the conversation goes with the results

from human being which was of course a lot of variation

because they were absent images

this is will be published in icp

we sat besides this doctoral dissertation this

there are there um many other details yeah in with these applications for the uh

forest industry but since i have limited time these i want to the selected so

so thank you very much

yeah

that i

i

it is difficult to the question and answer is that to my both are important

so actually what we finally would like to know is the volume of a of

those problems there

so then uh we need to detect the problems and but also distribution tells about

the process so about been experts can deal when the distribution changes

so both are unimportant and of course we make a quite a big assumption here

that uh if we think that this is a two dimensional bubble uh you meets

tells about the volume button roughly it goes right

in liquid

uh in this case would actually be are not and we um because these imaging

conditions are quite a typical bits totally inside the pipeline and some liquid is run

in there and we rely on that fact that at the place where you we

are sampling three presents the whole like with the area that might be a little

bit risky assumption but uh we uh we had second the possibility of having the

uh more images

uh

yeah

um

yeah we have a certain number of people first of all form in this uh

this uh a human soppy neon and then uh from their we tried to from

very low level attributes to build the higher level attributes

and the then uh koenig can make them to the physical measurements of is eight

samples of since we know each sample and we note no opinion of a human

being of from each sample and then uh then be uh for uh yeah

start to see the changes um actually it's a it's just send set us up

uh some kind of a curve but unfortunately i don't have the curve here we

very use it's like a rock or

and you check that how close you must goal that yeah this uh but utterance

order

close to the to the mean opinion score so that is the idea so it's

a little bit X experimental you have i understand your question but the that we

can say that it's a direct physical uh example exhumation which tells that very good

question

yeah this one

i

yeah

i

um

i actually um

uh we were of course we must have a normal each of these classes for

unsupervised and so what about what about what the band when checking the result of

an unsupervised uh

clustering but we have to do that without any knowledge of those classes

yeah so we have like we have about like uh database and there we have

exact knowledge of those categories and we at the end we check how it goes

so actually it's a huge the map where all these objects are clustered and then

we can start to see how it goes

and uh of course results are not so called then in a in a supervised

but we were quite surprised about one thing that is that the how a unique

the ballot the six people have used like something like uh having twenty of the

images and saying something about unsupervised clustering so we used hundreds of maybe six

for this

oh

i

actually we are trying to mister those things which affect the quality and actually they

affect the usability uh process is there

yeah

hmmm

well as well that's a of it that's a physical actually property of the paper

how we cepstral observed scene for example and that's quite the engineering style of testing

that there is really putting on the paper and then start into account where other

areas it up selves okay and we really doesn't to properly

yeah and before it happens that paper then yeah because actually we pretty there is

there are not so many technical problems anymore we have a somehow assume that

when the paper is okay it's not anymore so pick pretty the quality papers okay

ooh

i

uh excuse me what it what was

i mean how well you all these solutions might be what we said difficult to

say uh um then uh of course you can calculate the salary of those people

work in that club or authorities

and sela and the than the price of that uh

that solution that machine so that so quite easy to calculate but then when you

go to be clear peak approach to cease in paper making or impala been

that's a little bit more difficult to calculate but if you see that clear peak

lisa in the use of a raw material and water and unity i think the

estimates have been even up to forty percent

which is a lot of money