DE10235272A1 - Process for classifying and sorting hardwood lumber - Google Patents

Abstract

The invention relates to a method for classifying and sorting hardwood sawn timber, taking into account the dimensions and material properties. DOLLAR A According to the invention, each class can be designed on the basis of the following four parameters: DOLLAR A È total portion of the usable length lambda¶L¶: What minimum length portion must a faultless piece (cutting) have in the corresponding quality class. DOLLAR A È Total share of the usable width lambda¶B¶: What is the minimum width share of an error-free piece (cutting) in the corresponding quality class. DOLLAR A È Permissible number of useful items (cuttings) N: How often can the fault-free area be divided into the corresponding quality class. DOLLAR A È Type of optimization (longitudinal, cross or area cut): The maximum usable area (cutting) is preferably calculated in the longitudinal direction or in the cross direction or independent of direction for maximum area. DOLLAR A The method allows a clear, simple, fast, flexible and at the same time stable classification and sorting of hardwood lumber according to customer requirements and is suitable for both human and machine (e.g. with the help of an image processing system) classification and sorting.

Description

The invention relates to a method according to the generic term of claim 1.

In particular, the invention relates on the classification and sorting of hardwood lumber under consideration the measure and Material properties.

The sorting of lumber is a necessary requirement for an appropriate use of Wood. The goodness sorting determines the possible Use and is therefore of considerable economic importance for all Areas of wood processing and processing.

It is known that one for sorting of the pieces of wood (Boards, planks, etc.) needs a classification procedure, whereby the classes defined in advance according to appropriate criteria differ.

For Softwood lumber as a building material has been sorted according to Load capacity as standard procedure (DIN 4074-1). In doing so, everyone disorders the wood structure (defects) such as Knots, cracks, etc., which increase the strength of the wood, is taken into account.

When sorting hardwood So far, the sawn timber has been according to the optical Overall impression (Tegernsee customs) rated the errors, their condition and size, and their number and frequency the authoritative Evaluation criteria are.

In this process, which as a booth The technology is disadvantageous that only the detailed description the defect and its position are used as quality criteria, but not the flawless usable area authoritative for the Assessment of the sawn timber taken into account becomes. This method is therefore far from sufficient to achieve a clearly defined classification and thus a safe sorting to ensure.

Another procedure (rules for measuring & sorting North American Hardwood Lumber, National Hardwood Lumber Association, P.O.Box 34518, Memphis, Teenies 38184-0518, U.S.A.) used for Classification is a pattern that consists of defined classes exists, which are assigned to defined sections (cuttings). These sorting regulations are based mainly on the utilization of the sawn timber. This procedure is based on a comparison managed, on which the value of the boards can be determined.

Unfortunately, this procedure is over of complexity the class definition is very complex.

Besides, it doesn't allow any Length of Sawn timber and no free definition of the cuttings and thus of classes. So it is not flexible and explicit at the same time Classification and sorting of hardwood lumber according to customer requirements possible.

A similar solution is from EP 0899069A2 known in which a method and an apparatus for the further processing of unedged raw boards is described. With its help, the scrap can be reduced and the utilization of the raw material can be improved. The dimensions and all defects of the raw board are recorded. Then an individual cutting pattern is calculated for each raw board, in which imperfections are left out and the rest of the surface is divided over the different sections (scantlings) in such a way that optimal use of the material is achieved. The dimensions of the desired scantlings should be specified and specified.

However, this procedure does not define how the maximum yield of the raw board should be achieved and which parameters and limit values are responsible for this. There are no criteria for an objectively optimal way of occupying the usable area with error-free partial areas (Kanteln) have been called. This process remains completely subjective and therefore offers no possibility for one flexible and explicit classification and sorting of hardwood lumber.

The invention is based on the object a procedure for the classification and sorting of hardwood lumber to create a clear, simple, fast, flexible and at the same time explicit classification and sorting of hardwood lumber allowed according to customer requirements. The process is intended for both human for as well mechanical (e.g. with the help of an image processing system) classification and sorting.

The solution to the technical problem results from the features of claims 1 to 11.

According to the invention, the contours should be first and all defects in hardwood lumber are recognized and measured. Known for this Standards such as Tegernsee customs or DIN 68 367 "Determination the quality characteristics of hardwood lumber "and DIN 68 371 (pre-standard)" Measuring hardwood lumber " or features and dimensions specifically defined according to customer requirements become.

Normally up to a maximum of five grades can be used for hardwood lumber sorting Usually referred to as A, B, CD, E or as A, A / B, B, B / C, C. This is not intended to serve as a limitation of the proposed classification and sorting of hardwood lumber, but as an example.

According to claim 1, the identified Contours of the sawn timber, its dimensions (length L and width B) as well as size and position the errors found as starting data for the classification and sorting of hardwood lumber, with each class based on The following four parameters can be interpreted:

• - Total share of the usable length λ _L : What is the minimum length share of an error-free piece (cutting) in the corresponding quality class (in percent based on the total length: λ _L * 100%).
• - Total share of the usable width λ _B : What is the minimum width share of an error-free piece (cutting) in the corresponding quality class (in percent based on the total width: λ _B * 100%).
• - Permitted number of the useful pieces (Cuttings) N How often is the correct area in the corresponding quality class allowed patched become.
• - type of optimization (Along-, Cross or area cut): The maximum usable area (Cutting) is preferably done lengthways or calculated in the transverse direction or independent of direction for maximum area.

With the help of these four parameters different classes can be defined flexibly.

Furthermore, the dimensions of the absolute smallest possible Cuttings (length l abs / min and width b abs / min).

In order to be able to define a class, you first have to decide which type of optimization (longitudinal, cross-sectional or area cut) is preferred. Then the use of the length λ _L , the use of the width λ _B and the number of permissible cuttings N should be determined for each class.

If longitudinal section (LS) as an optimization type is preferred and the permissible Number of cuttings N greater than one is allowed the cuttings are in two rows next to each other for their maximum To reach length.

If cross-section (QS) as an optimization type is preferred and the permissible Number of cuttings N greater than one is, everyone should Cuttings lie one below the other to achieve their maximum width.

If area cut (FS) as optimization type is preferred and the permissible Number of cuttings N greater than one is allowed all Cuttings lie in pairs one below the other or next to each other in order to maximum area to reach.

Obviously the type of optimization is playing doesn't matter if the allowable Number of cuttings N is one.

It also doesn't make sense the permissible Number of cuttings N to be defined as three.

• – für Nⁱ = 1:
  
• – für Nⁱ = 2; Nⁱ ≥ 4: – beim Längsschnitt (LS):
  
  – beim Querschnitt (QS):
  
  – beim Flächenschnitt (FS):
  

wobei:
i – Klassenindex;
kⁱ – ganzzahliges Ergebnis der Division der Anzahl der klassenspezifischen zulässigen Cuttings Nⁱ mit zwei:

Anschließend werden die resultierenden minimalen Cutting-Größen noch mit den für alle Klassen geltenden vordefinierten absoluten minimalen Cutting-Größen verglichen. Für alle Klassen muss gelten:

Ansonsten sollen die klassenspezifischen Größen l i / min , b i / min durch die allgemein gültigen Größen l abs / min, b abs / min ersetzt werden. Damit sind die Klassen festgelegt.The four parameters mentioned above as well as the length L and width B of the current sawn timber are used to calculate the size of the class-specific minimum cuttings (length li / min and width bi / min). The following formulas should be used for this:

• - for N ⁱ = 1:
  
• - for N ⁱ = 2; N ⁱ ≥ 4: - in longitudinal section (LS):
  
  - for the cross-section (QS):
  
  - for area cutting (FS):
  

in which:
i - class index;
k ⁱ - integer result of dividing the number of class-specific permissible cuttings N ⁱ by two:

Then the resulting minimum cutting sizes are compared with the pre-defined absolute minimum cutting sizes that apply to all classes. The following must apply to all classes:

Otherwise, the class-specific sizes li / min, bi / min should be replaced by the generally applicable sizes l abs / min, b abs / min. The classes are now defined.

With further sorting should on the lumber maximum possible error-free areas (cuttings) taking into account the type of optimization determined and their size be sorted. Then may only the first N allowed Cuttings are used to define the class, if these available and not less than the class-specific minimum Cuttings are. It should be noted that the number N per class is designed. Surplus cuttings do not go into the calculation of the total usable area for the corresponding Class one.

All classes should be checked until a class turns out to be valid proves.

wobei:
i – Klassenindex;
j – Cuttingsindex;
Nⁱ – Anzahl der für i-te Klasse zulässigen Cuttings;
l_j – Länge der j-ten Cutting;
b_j – Breite der j-ten Cutting; Die ermittelte Ausbeute γⁱ wird mit einer klassenspezifischen minimal erforderlichen Ausbeute

verglichen. Sobald: γ ≥ γimin erweist sich eine Klasse als passend, und es findet keine weitere Überprüfung mehr statt.The area of the cuttings valid for each class should be summed up. With this amount, a class-specific yield γ ^{i is} calculated as a proportion of the total area:

in which:
i - class index;
j - cutting index;
N ⁱ - number of cuttings permitted for i class;
l _j - length of the j-th cutting;
b _j - width of the j-th cutting; The determined yield γ ⁱ becomes with a class-specific minimum required yield

compared. As soon as: γ ≥ γ i min a class turns out to be suitable and no further verification takes place.

This makes the process flexible, simple and allows quick and explicit classification and sorting of hardwood lumber. The minimum and determined yield of the sawn timber for the determined class γ i / min, γ ⁱ (in percent of the total area: γ i / min * 100%, γ ⁱ * 100%) can be made available as additional information.

As is known, the width of the sawn timber for untrimmed sawn timber can be determined according to different types (Tegernsee customs): half-tree width, top or cut width. With trimmed sawn timber, of course, only the cutting width can be measured. Therefore, the total proportion of the usable width depends on the associated type of determination and allows timber to be classified and sorted accordingly. According to claim 2, the type of determination of the width can be taken into account when evaluating the class and thus allows a dependent classification and sorting of hardwood lumber. Depending on the type of definition, the width of the sawn timber B should be determined as a half-tree width B _HB , as a deck width B _D or as a cutting width B _S and taken into account when determining the class and further sorting.

According to claim 3 can each one to be checked on the wood Object such as Branch, crack, red core etc. or even wood structure depending on the type, size or Size either as an error or as a general characteristic of the wood (no Errors) and can handle the classification and Sorting of sawn timber influence. The errors determine the cuttings and thus the corresponding ones Grade. The presence of features can have different effects the classification and sorting of hardwood lumber. The characteristics can either be taken into account immediately when generating the cuttings or only first as an additional quality criterion for already created cuttings serve. You can ascertained quality classes improved or worsened by one or more levels or even an extra defined class can be created. For example The presence of the red kernel can be classified into an extra Leading class.

According to claim 4 can the type of error and its maximum size that is allowed when generating the Cuttings considered should be determined per class. That for example, that different Error types and error size for different ones Classes decisive are. The classification and sorting of hardwood lumber wins greater flexibility.

According to claim 5 can the type of optimization that is taken into account when generating the cuttings should determine per class. That for example, that for a higher class only longitudinal section, for a lower Great area cut may be allowed. The classification and sorting of Hardwood sawn timber greater flexibility.

According to claim 6 can the absolutely minimum size of the cuttings (Length l abs / min and width b abs / min) per class: l i_abs / min, b i_abs / min. So that wins Classification and sorting of hardwood lumber gives greater flexibility.

According to claim 7, the cuttings required for a class can be defined not only as a percentage but also absolutely according to size, with the minimum length li / min and width b / min of an error-free piece (cutting), which may occur in the corresponding quality class, being defined directly can be. The class-specific yield γ ^{i should} be calculated as a proportion of the total area.

verglichen. Sobald: γ ≥ γimin erweist sich eine Klasse als passend, und es findet keine weitere Überprüfung mehr statt.The yield γ ⁱ determined is obtained with the following, as defined in claim 1, class-specific minimum required yield

compared. As soon as: γ ≥ γ i min a class turns out to be suitable and no further verification takes place.

In this case, the general remain or class-specific absolute minimum cutting sizes l abs / min, b abs / min and l i_abs / min, b i_abs / min disregarded.

According to claim 8, the cuttings required for a class can be defined absolutely in terms of size, but the class-specific minimum number of permissible cuttings N i / min should serve as a necessary criterion for class definition. This means that at least the required number of N i / min permissible cuttings should be taken out of the sawn timber. As soon as: N i ≥ N i min, a class turns out to be suitable and no further verification takes place.

In this case, the general remain or class-specific absolute minimum cutting sizes l abs / min, b abs / min and l i_abs / min, b i_abs / min and the class-specific minimum yield γ i / min are not taken into account.

As is known, hardwood lumber is available in various lengths. A standard length L ₀ of hardwood lumber, to which the class-determining parameters are based, can be specified according to different standards, e.g. 1m (DIN 68 367, DIN 4074) or 3m (Tegernsee customs). Nevertheless a longer piece of wood should be treated just like one of the same quality but shorter.

Logically, the class-related parameters such as the total share of the usable length λ _L , the total share of the usable width λ _B and the type of optimization (longitudinal, cross or surface section) must not be changed. The only thing that changes is the permissible number of cuttings N. According to claim 9, the class-specific permissible number of cuttings N ⁱ (L ₀ ), which was designed for a standard length L ₀ , is to be converted for a real length L, if this exceeds the standard length L ₀ . The new class-specific permissible number of cuttings N ⁱ (L) should be calculated as follows:

• - for N ⁱ (L ₀ ) = 1: N i (L) = N i (L 0 ) * p = p ;
• - for N '= 2; N 'Z 4: N i (L) = N i (L 0 ) * p + 2 * q i if: η <η i min (L) ; otherwise: N i (L) = N i (L 0 ) * p + 2 * q i + 1 , where: N ⁱ (L ₀ ) - class-specific permissible number of cuttings on the standard length L ₀ ; p - integer result of dividing the length of the timber with the standard length L ₀ :
  
  q ⁱ - integer result of dividing the remaining length Δl of the sawn timber by the class-specific local standard length l ₀ :
  
  η - the ratio between the real length of the sawn timber L and the standard length L ₀ :
  
  η i / min (L) - a class and length-specific minimum limit ratio.

The remaining length Δl, the class-specific local standard length l ₀ and the class-specific minimum limit ratio η i / min (L) are: Δl = L - L 0 * p ;

As is known, sawn timber can have different defects on the different sides, which also occur in different places. According to claim 10, both the best and the worst side and the two sides combined as a so-called image of the sawn timber can be taken into account in the classification and sorting. When classifying the better or worse side, each side of the sawn timber is classified independently of one another. Then the better or the worse side is used as the basis for evaluating the sawn timber and, if desired, the other side is used for correction. For example, a lumber can be evaluated on the better side, whereby the worse side could cause a gradation by one or more levels. In the opposite case, a lumber can be evaluated on the worse side, whereby the better side could cause a ranking by one or more levels.

The screen represents all errors and edges of the sawn timber on both sides and thus allows a generation of the cuttings that are error-free on all sides. This strictest type of evaluation brings the smallest yield but for that the best quality of goods.

According to claim 11 can the classification and sorting of hardwood lumber for commercial purposes considering the for the individual classes are valid Carry out prices. This means that lumber that has been classified in a higher class but a greater yield in a lower class and thus has a larger value, preferred be classified in the lower class.

Thus, both classification, which only refers to defects but not to the faultless usable area (Tegernsee customs), as well as the complex and inflexible classification of hardwood lumber (Rules for the measurement & sorting of North American hardwood lumber) by the here mentioned Process for classifying and sorting hardwood lumber to be dispensed with.

The details of the invention are described in the following exemplary embodiments 1 - 7 and Tables 1-3 explained.

A four-level classification can be used as an example. For these, classes A, B, C and D can be defined as shown in Table 1. The total share of the usable length λ _L and the total share of the usable width λ _B as well as the type of optimization (longitudinal section) for the quality classes A, B and C are the same. Class D is for the rest. The only parameter that changes from one grade to another is the permissible number of cuttings N ⁱ .

For the highest class A ( 1a ) may only be a complete piece of the lumber to be tested (cutting) 2 come out (N ⁱ = 1). If this piece has a yield γ ^A that is greater than the class-specific minimum yield γ A / min (γ ^A ≥ γ A / min), the lumber is recognized as good and is classified in class A.

The lumber that contains a flaw in the middle no longer allows a single flawless piece ( 1b ). That is taken into account by two pieces 2 from the lumber to be tested 1 can be used (N ⁱ = 2). If these pieces together have a yield γ ^B that is greater than the class-specific minimum yield γ B / min (γ ^B ≥ γ B / min), the sawn timber is recognized as good, but is classified in a quality class lower, namely in class B. ,

If there are several defects on the sawn timber 1 but four largest possible flawless pieces occurred 2 together can still guarantee the necessary class-specific minimum yield γ C / min (γ ^C ≥ γ C / min) ( 1 . c ), this lumber is classified in quality class C.

All other cut woods that not in these three grades fit, should be classified in class D (worst product).

For the other types of optimization (cross-section and area cut) with the same parameters (Table 1), there are corresponding examples in 2 and 3 shown.

Table 1.

According to the patent, the class-specific minimum cutting sizes (li / min, bi / min) should also be compared with the absolute minimum cutting sizes (l abs / min, b abs / min) that apply to all classes. A distinction is made between the largest possible (li / max, bi / max) and the smallest possible (li / min, bi / min) classes within each class specific cutting. These two cuttings form a class-specific pair that occupies the k ⁱ part of the sawn timber and has the k ⁱ part of the required error-free area, where: k ⁱ - integer result of dividing the number of class-specific permissible cuttings N ⁱ by two :

You can see the size of the smallest possible Cuttings based on length L and the width B of the lumber as well as the size of the largest possible Calculate cuttings.

For N ⁱ = 1 these cuttings are identical:

For N ⁱ ≥ 2, they depend on the type of optimization.

The dimensions of the permissible cuttings should be calculated from a pair of the minimum and maximum cuttings. Since the sawn timber has a much larger dimension in the longitudinal direction than in the transverse direction, the further increase in the permissible number of cuttings only affects the minimum length, but not the minimum width of the cuttings. Therefore, the usable width should only be guaranteed with a maximum of two cuttings (LS, FS) or one cutting (QS, FS). The maximum cutting takes up half the area of the k ⁱ th part of the sawn timber:

With the longitudinal cut (LS) one gets for a number of the permissible cuttings N ⁱ ( 4 . a ):

wobei:
kⁱ – Anzahl der größtmöglichen Cuttings,
(Nⁱ – kⁱ) – Anzahl der kleinstmöglichen Cuttings.The total usable length for this piece to be examined 1 taking into account the double length, which consists of two rows of adjacent cuttings 2 arises, is:

in which:
k ⁱ - number of the largest possible cuttings,
(N ⁱ - k ⁱ ) - Number of smallest possible cuttings.

From (4) and (2a) we get:

It is easy to see that the values of li / min LS are the same for N ⁱ = 4 and N ⁱ = 3:

In contrast, the greatest possible cutting for N ⁱ = 3 should be significantly larger than for N ⁱ = 4 ( 5 ). In order to optimally assign the lumber to be examined 1 with reasonable cuttings 2 to achieve it is therefore not sensible to define the permissible number of cuttings N ⁱ as three.

For the entire usable width you get under the same conditions:

From (8) and (4b) we get:

With the cross-section (QS) one gets for a given number of permissible cuts N ⁱ ( 4 . b ):

The total usable length for the lumber to be examined 1 taking into account the simple length that results from a series of cuttings lying one below the other 2 arises, is:

From (11) and (10a) we get:

For the entire usable width you get under the same conditions as for the longitudinal cut:

From (13) and (10b) we get:

Because the surface section (FS) includes both the longitudinal section and the cross section, the associated minimum permitted cutting should have the minimum length of the cross section and the minimum width of the longitudinal section:

In the case of cross-section and area cutting as well as in longitudinal section, it does not make sense to define the permissible number of cuts N ⁱ as three.

For all cases and for all classes: λ i L >0.50; (17a) λ i B > 0.50, (17b) what is practical and

Otherwise, the class-specific Sizes l i / min, b i / min through the generally applicable Sizes l abs / min, b abs / min replaced become.

As a further example, a five-level classification can be assumed, whereby the type of optimization is always defined as a longitudinal section, but the other parameters vary for the different classes (Table 2, 6 ).

Table 2.

All of the above examples relate to the case in which the length of the sawn timber L does not exceed a predefined standard length L ₀ . Otherwise, the class-specific permissible number of cuttings N ⁱ (L ₀ ), which is designed for the standard length L ₀ , should be replaced by a new value N ⁱ (L) corresponding to the real length of the sawn timber L.

If the length of the sawn timber L is a multiple of the standard length L ₀ ( 7 . a ), the number N ⁱ (L ₀ ) should increase by p times, where p is the integer result of dividing the length of the sawn timber L by the standard length L ₀ :

ergibt sich: Ni(L) = Ni(L0)*p (21a) oder, falls Nⁱ(L₀) = 1: Ni(L) = p. (21b)For an integer value of the ratio

surrendered: N i (L) = N i (L 0 ) * P (21a) or, if N ⁱ (L ₀ ) = 1: N i (L) = p. (21b)

For N ⁱ (L ₀ ) = 1, (21b) is valid for all other η.

For N ⁱ (L ₀ )) = 2 or N ⁱ (L ₀ ) ≥ 4, there may be further cuttings on the remaining length Δl if the ratio η is not an integer but a value greater than one,
in which: Δl = L - L 0 * p. (22)

berechnet werden. Diese Länge entspricht dem Bereich, auf dem sich exakt ein Paar von Cuttings befinden kann (7,b).This additional number of cuttings can be made using a class-specific local standard length

be calculated. This length corresponds to the area on which exactly one pair of cuttings can be located ( 7 . b ).

This means that there can still be q ⁱ cutting pairs on the remaining length Δl, where q ^{i is} the integer class-specific result of dividing the remaining length Δl of the sawn timber by the class-specific local standard length l ₀ :

For an integer value of q ⁱ : N i (L) = N i (L 0 ) * p + 2 * q i , (25)

If the length L is increased further, there is a length L ~ which can accommodate the smallest possible cutting first ( 7 . e ): L ~ = L 0 * p + 2 * q i + 1 (26) and N i (L ~) = N i (L 0 ) * p + 2 * q i + 1. (27)

The total usable length should match the cuttings achieved. With the longitudinal cut (LS), taking into account the double length 2 * L ~, which results from two rows of cuttings next to each other, you get an entire usable length:

From (28) with (6) and (4a) with L ₀ instead of L we get:

A class-specific and length-specific minimum limit ratio η i / min_LS (L) can be calculated from the longitudinal section:

The coefficients η and η i / LS (L) can be used to decide how the new class-specific permissible number of cuttings N ⁱ (L) should be calculated. If η <η i / min_LS (L), the number N ⁱ (L) is calculated according to (25), otherwise according to (27).

Im Fall p = 1 kann (31) folgendermaßen berechnet werden:The same procedure can be used for cross-sectional and surface cutting with the same result:

In the case of p = 1, (31) can be calculated as follows:

• - for just class-specific permissible number of cuttings N ⁱ (L ₀ ) = 2 * k ⁱ
  
• - for odd class-specific permissible number of cuttings N ⁱ (L ₀ ) = 2 * k ⁱ + 1:
  

As an example, the new class-specific permissible number of cuttings N ⁱ (L) for the three classes of the above-mentioned four-stage classification (longitudinal section), which was designed for a standard length L ₀ = 3.0m, over a real length L of 3.0m to 6.0m (with an accuracy of 0.1m) (Table 3).

Table 3.

The corresponding tables, such as Table 1-3, you can for any both class and also create process-specific parameters with the help of (1) to (30).

This allows a clear, simple, fast, flexible and at the same time explicit classification and Sorting hardwood lumber according to customer requirements both human as well as mechanically (e.g. with the help of an image processing system) carry out.

1

Lumber

2

flawless piece of wood (Cutting)

Claims (11)

Procedure for the classification and sorting of hardwood sawn timber taking into account the dimensions and material properties, characterized in that the identified contours of the sawn timber, the dimensions (length L and width B) as well as the size and position of the errors found are used as starting data for the classification and sorting of hardwood sawn timber should be designed, whereby each class can be designed on the basis of the following four parameters: - Total portion of the usable length λ _L : What is the minimum length portion of an error-free piece (cutting) in the corresponding quality class. - Total share of the usable width λ _B : What is the minimum width share of an error-free piece (cutting) in the corresponding quality class. - Permissible number of useful items (cuttings) N How often can the fault-free area be divided into the corresponding quality class. - Type of optimization (longitudinal, cross or area cut): The maximum usable area (cutting) is preferably calculated in the longitudinal direction or in the cross direction or independent of direction for maximum area. Furthermore, the dimensions of the absolutely smallest possible cuttings (length l abs / min and width b abs / min) have to be determined. According to the patent, the classification and sorting of hardwood lumber is to be carried out as follows. First, it should be defined which type of optimization (longitudinal, cross or area cut) is preferred, with the longitudinal cut (LS) the cuttings may lie in two rows next to each other, with the cross section (QS) all cuttings should be underneath each other and with the area cut (FS) All cuttings may lie in pairs one below the other or next to each other. Then the use of the length λ _L , the use of the width λ _B and the number of permissible cuttings N should be determined for each class. Next, the size of the class-specific minimum cuttings (length li / min and width bi / min) are calculated using the following formulas: - for N ⁱ = 1: l i min = L * λ i L ; b i min = L * λ i B ; - for N ⁱ = 2; N ⁱ ≥ 4: - in longitudinal section (LS):

- for the cross-section (QS):

- for area cutting (FS):

where: i - class index; k ⁱ - integer result of dividing the number of class-specific permissible cuttings by two:

The resulting minimum cutting sizes are then compared with the pre-defined absolute minimum cutting sizes that apply to all classes. The following must apply to all classes:

Otherwise, the class-specific sizes li / min, bi / min should be replaced by the generally applicable sizes l abs / min, b abs / min. The classes are now defined. With further sorting, the maximum possible error-free areas (cuttings) should be determined on the sawn timber, taking into account the type of optimization, and sorted according to their size. Then only the first N permissible cuttings may be used to define the class, if these exist and are not smaller than the class-specific minimum cuttings. It should be noted that the number N per class is designed. Surplus cuttings are not included in the calculation of the total usable area for the corresponding class. All classes should be checked until one class proves to be valid. The area of the cuttings valid for each class should be summed up. With this amount, a class-specific yield γ ^{i is} calculated as a proportion of the total area:

where: i - class index; j - cutting index; N ⁱ - number of cuttings permitted for i-th class; l _j - length of the j-th cutting; b _j - width of the j-th cutting. The determined yield γ ⁱ becomes with a class-specific minimum required yield γ i min = λ i L * λ i B compared. As soon as: γ ⁱ ≥ γ i / min a class proves to be suitable and no further verification takes place. Method according to Claim 1, characterized in that the type of determination of the width is taken into account in the class determination and further sorting, in that the width of the sawn timber B is determined depending on the type of definition, as a half-tree width B _HB , as a deck width B _D or as a cutting width B _S. A method according to claim 1, characterized in that each to be tested on the wood Object such as Branch, crack, red core etc. or even wood structure depending on the type, size or Size yourself either as a flaw or as a general characteristic of the wood can be treated and accordingly influences the classification and sorting of sawn timber. The errors determine the cuttings and thus the quality class. The presence of features can either be the same when the cuttin is created considered as a quality criterion or only with the already created cuttings serve. You can already determined quality classes improved or worsened by one or more levels or even an extra defined class can be created. A method according to claim 1, characterized in that the type of error and its maximum size allowed when generating the Cuttings considered can be determined per class. A method according to claim 1, characterized in that The type of optimization that takes into account when generating the cuttings can be determined per class. A method according to claim 1, characterized in that the absolutely minimum size of the cuttings (Length l abs / min and width b abs / min) per class (l i_abs / min, b i_abs / min.). A method according to claim 1, characterized in that the cuttings required for a class can be defined not only as a percentage, but also absolutely according to size, the minimum length li / min and width bpm of an error-free piece (cutting), which may occur in the corresponding quality class, can be defined directly. The class-specific yield γ ^{i should} be calculated as a proportion of the total area and with the following class-specific minimum required yield

be compared. As soon as: γ i ≥ γ i min a class turns out to be suitable and no further verification takes place. Method according to claim 7, characterized in that the class-specific minimum number of permissible cuttings N i / min is to serve as a necessary criterion for the class definition, the cuttings required for a class being defined absolutely according to size. This minimum required number of N i / min permissible cuttings should be able to be extracted from the sawn timber. As soon as: N i ≥ N i min , a class turns out to be suitable and no further verification takes place. Method according to claim 1, characterized in that the class-specific permissible number of cuttings N ⁱ (L ₀ ), which has been designed for a standard length L ₀ , is to be converted for a real length L if this exceeds the standard length L ₀ . The new class-specific permissible number of cuttings N ⁱ (L) should be calculated as follows: - for N ⁱ (L ₀ ) = 1: N i (L) = N i (L 0 ) * p = p; - for N ⁱ = 2; N ⁱ ≥ 4: N i (L) = N i (L 0 ) * p + 2 * q i . if: η <η i min (L) otherwise: N i (L) = N i (L 0 ) * (p + 1) + 2 * q i +1, where: N ⁱ (L ₀ ) - class-specific permissible number of cuttings on the standard length L ₀ ; p - integer result of dividing the length of the timber with the standard length L ₀ :

q ⁱ - integer result of dividing the remaining length Δl of the sawn timber by the class-specific local standard length l ₀ :

η - the ratio between the real length of the sawn timber L and the standard length L ₀ :

η i / min (L) - a class and length-specific minimum limit ratio. The remaining length Δl, the class-specific local standard length l ₀ and the class-specific minimum limit ratio η i / min (L) are: Δl = L - L 0 * p;

A method according to claim 1, characterized in that both the best and the worst side as well as the two Pages combined as a so-called image of the sawn timber classification and sorting. In the Classification according to better or worse side should each side of the sawn timber is independently classified become. One side is used as the basis for evaluating the sawn timber taken and the other, if desired, to correct the evaluation consulted and the classification of the sawn timber by one or corrected several levels down or up. The image represents all defects and edges of the sawn timber on both sides and thus allows the cuttings to be generated from all sides are flawless. A method according to claim 1, characterized in that the classification and sorting of hardwood lumber taking into account the for the individual classes are valid Carry out prices leaves, by using a lumber that has been classified in a higher class but a greater yield in a lower class and thus has a larger value, preferred can be classified in the lower class.