Invention content
The object of the present invention is to provide a kind of sawn timber surface quality grades to divide device and method, improves classification accuracy.
To achieve the above object, the present invention provides following schemes:
A kind of sawn timber surface quality grades divide device, including:Moisture content is disposed with along the direction of transfer of sawn timber
Detection device, weighing sensor, vision inspection apparatus and code-spraying equipment;The detecting device for moisture content, the weighing sensor,
The output terminal of the vision inspection apparatus with middle control mechatronics;The detecting device for moisture content is used to detect the sawn timber
Moisture content;The weighing sensor is used to detect the weight of the sawn timber;The vision inspection apparatus is used to detect the sawn timber
Surface defect and identify sawn timber annual ring;The code-spraying equipment is used to stamp table to the sawn timber under the control of the middle control machine
Face credit rating and the stress levels of entry evaluation label;
The vision inspection apparatus includes six positive shooting cameras, and six cameras are being separately positioned on position to be captured just
On top, underface, directly to the left, directly to the right, front and dead astern six direction.
Optionally, the vision inspection apparatus further includes multiple oblique shooting cameras, and multiple oblique shooting cameras are located at described treat
The oblique upper and obliquely downward of camera site;Each shooting angle of the oblique shooting camera and the angle of horizontal plane 20 °~70 ° it
Between it is adjustable.
Optionally, which further includes the first photoelectric sensor, the second photoelectric sensor, third photoelectric sensing
Device and the 4th photoelectric sensor;
First photoelectric sensor is between the conveyer belt initial position and the detecting device for moisture content
And close to the detecting device for moisture content;Second photoelectric sensor is located at the detecting device for moisture content and the biography of weighing
Between sensor and close to the weighing sensor;The third photoelectric sensor is located at the weighing sensor and is examined with the vision
It surveys between device and close to the vision inspection apparatus;4th photoelectric sensor be located at the vision inspection apparatus with it is described
Between code-spraying equipment and close to the code-spraying equipment.
Optionally, the vision inspection apparatus is arranged in Image Acquisition room.
Optionally, it is additionally provided with lighting device in described image collection room.
Optionally, which further includes motor;The motor drives the sawn timber to pass for driving conveyer belt
It send.
The invention also discloses a kind of sawn timber surface quality grades division methods, applied to above-mentioned sawn timber surface quality etc.
Grade divides device;This method includes:
Obtain the moisture content of the detecting device for moisture content detection;
Obtain the weight of the weighing sensor detection;
Obtain the image of the vision inspection apparatus shooting;
Type, profile, sawn timber annual ring and the surface defect of the sawn timber are identified according to described image;
The volume of the sawn timber is calculated according to the profile;
The density of the sawn timber is calculated according to the volume and the weight;
Sawn timber surface quality grades are divided according to the surface defect;
It will be in the density, the type, the sawn timber annual ring, the surface defect and the moisture content and database
Data are compared, and assess the stress levels of the sawn timber;
The code-spraying equipment is controlled to stamp surface quality grades label to the sawn timber according to the surface quality grades;
The code-spraying equipment is controlled to stamp stress levels label to the sawn timber according to the stress levels.
According to specific embodiment provided by the invention, the invention discloses following technique effects:Sawn timber disclosed by the invention
Surface quality grades divide device and method, and whether the surface using vision inspection apparatus analysis sawn timber meets preset condition, from
And range estimation classification is carried out to sawn timber using machine vision technique, improve accuracy and the efficiency of sawn timber range estimation hierarchical detection.Together
When, it using the volume and type of image analysis sawn timber, and identifies sawn timber annual ring, quality is measured using weighing sensor, with reference to saw
The volume of material is so as to obtain the density of sawn timber, according to density, the type of sawn timber, annual ring information, moisture content, surface defect, combination
Sawn timber density in experimental data base in middle control machine, the type of sawn timber, annual ring information, moisture content, the elasticity corresponding to surface defect
Modulus gives tested sawn timber one preliminary stress grading, and the sawn timber to be classified without mechanical stress provides preliminary stress grading
With reference to.Moreover, the present invention realizes multiple device continuous actions, the automation of lumber class division is realized, improves sawn timber etc.
The efficiency that grade divides.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without making creative work
Embodiment shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, it is below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
Fig. 1 is the structure drawing of device that sawn timber surface quality grades of the present invention divide device embodiment one.
Referring to Fig. 1, which divides device device, including:Rack 1, conveyer belt 2, motor 3.It is described
Conveyer belt 2 is arranged on the rack 1, and the motor 3 drives the sawn timber 4 to transmit for the conveyer belt 2 to be driven to transmit.
Detecting device for moisture content 5, weighing sensor 6, vision inspection apparatus are disposed with along the direction of transfer of sawn timber 4
7 and code-spraying equipment 8;The detecting device for moisture content 5, the weighing sensor 6, the output terminal of the vision inspection apparatus 7 are equal
With middle control mechatronics;The detecting device for moisture content 5 is used to detect the moisture content of the sawn timber 4;The weighing sensor 6 is used
In the weight for detecting the sawn timber 4;The vision inspection apparatus 7 is used to detect described in surface defect and the identification of the sawn timber 4
The annual ring of sawn timber 4;The code-spraying equipment 8 is used to stamp grade mark to the sawn timber 4 under the control of the middle control machine;It is described
Vision inspection apparatus 7 is arranged in Image Acquisition room 9;Lighting device 10 is additionally provided in described image collection room 9.
The grade classification device further includes the first photoelectric sensor 11, the second photoelectric sensor 12, third photoelectric sensor
13 and third photoelectric sensor 14;First photoelectric sensor 11 be located at the conveyer belt initial position with it is described aqueous
Between rate detection device 5 and close to the detecting device for moisture content 5, whether enter moisture content detection for detecting the sawn timber 4
Region;Second photoelectric sensor 12 is between the detecting device for moisture content 5 and the weighing sensor 6 and close to institute
Weighing sensor 6 is stated, for detecting whether the sawn timber 4 enters weighing area;The third photoelectric sensor 13 is located at described
Between weighing sensor 6 and the vision inspection apparatus 7 and close to the vision inspection apparatus 7, it is for detecting the sawn timber 4
No vision-based detection region of entering;The third photoelectric sensor 14 be located at the vision inspection apparatus 7 and the code-spraying equipment 8 it
Between and close to the code-spraying equipment 8, for detecting whether the sawn timber 4 enters marked region.First photoelectric sensor 11, second
Photoelectric sensor 12, third photoelectric sensor 13 and third photoelectric sensor 14 are respectively detecting device for moisture content 5, weighing and sensing
Device 6, vision inspection apparatus 7 and code-spraying equipment 8 provide enabling signal.
Fig. 2 is the apparatus structure for the vision inspection apparatus 7 that sawn timber surface quality grades of the present invention divide device embodiment one
Figure.
Referring to Fig. 2, which includes six positive shooting cameras, and six cameras are separately positioned on to be captured
On the surface of position, underface, directly to the left, directly to the right, front and dead astern six direction.
Six positive shooting cameras mainly for detection of the middle section of sawn timber 4, fasciation section, combination section, knothole, small worm channel, grub hole,
The defects of being machined offset, skips-in-planting, splitting, rotten, cross grain.These defects are all happened at the wide face of sawn timber and narrow substantially
Face and the end position of sawn timber, it is only necessary to arrange that four cameras in wide face and leptoprosopy and both ends arrange that two cameras carry out image
Acquisition.
Fig. 3 is the apparatus structure for the vision inspection apparatus 7 that sawn timber surface quality grades of the present invention divide device embodiment two
Figure.
Referring to Fig. 3, which further includes multiple oblique shooting cameras, and multiple oblique shooting cameras are located at described treat
The oblique upper and obliquely downward of camera site;Each shooting angle of the oblique shooting camera and the angle of horizontal plane 20 °~70 ° it
Between it is adjustable.
Oblique shooting camera hinders hole, saw kerf, wane, crackle, three faces section, strip section, round edge mainly for detection of the processing of sawn timber
These defects such as section, blue change.These defects mainly appear on the corner of the side of sawn timber and on two faces of sawn timber width.
The present invention can realize the identification of sawn timber type.Sawn timber during growth, understands self-assembling formation native texture, day
Right texture shows certain systematicness, and longitudinal stripe texture is presented in some, and transverse pattern texture is presented in some, and the present invention is by setting
The native texture regular nature that multiple cameras can be good at detecting sawn timber is put, there is sawn timber texture depth identification function.Sawn timber
Texture contrast, which has deeply, has shallow, deeper texture and shallower texture that can well detected by the present invention;Different
Sawn timber has different twills, and the fractal characteristic of sawn timber radial direction, tangential texture is visited by using fractal geometry analytic approach
It begs for, the values of fractal dimension of sawn timber texture characterizes the shape of sawn timber surface texture, distribution density, uniformity coefficient and width characteristics,
And then judge the type of detected sawn timber.
The present invention can realize that defect kind identifies, detailed process is:
Using neural network algorithm, research is identified to the type of sawn timber surface defect:Sawn timber surface defect has ash
The characteristics of spending difference, shape difference, disparity, the present invention select gray feature, shape feature, geometric properties to establish feature
Database, as the basis for estimation of defect kind, the process for establishing database is:
The basic configuration of defect is portrayed using minimum enclosed rectangle:According to the length and width ratio of minimum enclosed rectangle as special
Value indicative distinguishes leptosomatic defect (strip section, basic wane, crackle);Saw kerf, worm channel, knothole are distinguished according to rectangle angle value
These three defects;
Reflect the location information of defect by establishing coordinate system, available for middle section, fasciation section, combination section, knothole, small worm
The defects of hole, grub hole, mechanical processing offset, skips-in-planting, splitting, rotten, cross grain.These defects are all happened at sawn timber substantially
Wide face and leptoprosopy, abscissa (coordinate of sawn timber axial direction) has good effect for distinguishing these defects, and ordinate (is sawed
The coordinate of material radial direction) it is little for distinguishing defect type effect, it does not need to use.
The gray feature of average gray value and gray scale mean square deviation reflection defect can reflect the bright dark feature of defect, die for the sake of honour,
Movable joint, occur the blue sawn timber average gray value for becoming part and gray scale mean square deviation have apparent otherness, established database it
Afterwards, by being analyzed and processed to the image collected, determine that the minimum enclosed rectangle of defect, the area of defect and minimum are external
The area of rectangle considers factors above pair than, the abscissa of defect and the average gray value of defect and gray scale mean square deviation
Defect to be identified carries out defect kind identification, and detailed process is:
The feature of said extracted is handled according to the present invention, selects 7 characteristic values as neural network to sawn timber table
The classification indicators of planar defect, including breadth length ratio, direction offset, abscissa account for overall length ratio, rectangular degree, consistency logarithm,
Average gray value accounts for tonal gradation ratio, gray scale mean square deviation, is formed 7 indexs for judging sawn timber surface defect.It will wait to know
Other 19 kinds of defects are as output of the neural network to judging result.
The type of defect is obtained according to system to the calculating of seven characteristic values and with desired value comparative analysis.
The present invention can realize the identification of sawn timber annual ring, specially:For tree ring feature of image, wavelet transformation is utilized
Method the data of annual ring image are compressed, under the premise of higher resolution is ensured, obtain higher compression ratio, then
Annual ring image is split, analyzes and finally identifies the sawn timber annual ring.
The present invention can also judge whether the knot on sawn timber surface meets preset condition, and detailed process is:Obtain the saw
The image on material surface;The knot on the sawn timber surface is identified using image recognition technology;Judge the position of the knot, size and
Whether quantity meets preset value, when any one in the position of the knot, size and number does not meet preset value, determines
Sawn timber is unqualified sawn timber.
The present invention can also judge whether the latewood percentage of sawn timber meets latewood percentage preset condition, and detailed process is:
Using texture recognition, late wood's fringe region in the growth ring and each growth ring on the sawn timber surface is identified;
Calculate the width of the growth ring and the width of late wood in each growth ring;
Latewood percentage is calculated using the following formula:
In formula:LwThe latewood percentage of sample,
ΣLbLate wood's overall width in measurement range, unit are millimeter
Wheel overall width in the whole year in b-measurement range, unit is millimeter
By the latewood percentage and default latewood percentage threshold value comparison, when the latewood percentage exceeds the default latewood percentage threshold value
When, determine that the latewood percentage does not meet preset condition, it is unqualified sawn timber to determine the sawn timber;Described in meeting when the latewood percentage
During default latewood percentage threshold value, determine that the latewood percentage meets preset condition, it is qualified sawn timber to determine the sawn timber.
The present invention can also judge whether the blue control with changed scale of sawn timber meets blue change preset condition, and detailed process is:
Identify that the blue region become occurs for the sawn timber surface;
Calculate the area in region and the sawn timber gross area that blue change occurs;
The blue control with changed scale is calculated according to the gross area of the area that the blue region become occurs and the sawn timber;
The blue control with changed scale and default blue control with changed scale threshold value are compared, when the blue control with changed scale is beyond the default blue control with changed scale
During threshold value, it is unqualified sawn timber to determine the sawn timber;When the blue control with changed scale meets the default blue control with changed scale threshold value, determine
The sawn timber is qualified sawn timber.
Fig. 4 is the method flow diagram of sawn timber surface quality grades division methods embodiment of the present invention.
Referring to Fig. 4, which divides dress applied to above-mentioned sawn timber surface quality grades
It puts;This method includes:
Step 401:Obtain the moisture content that the detecting device for moisture content 5 detects;
Step 402:Obtain the weight that the weighing sensor 6 detects;
Step 403:Obtain the image that the vision inspection apparatus 7 is shot;
Step 404:Type, profile and the surface defect of the sawn timber 4 are identified according to described image;
Step 405:The volume of the sawn timber 4 is calculated according to the profile;
Step 406:The density of the sawn timber 4 is calculated according to the volume and the weight;
Step 407:The surface quality grades of the sawn timber 4 are divided according to the surface defect;
Step 408:By the density, the type, the sawn timber annual ring, the surface defect and the moisture content and number
It is compared according to the data in library, assesses the stress levels of the sawn timber;
Step 409:The code-spraying equipment is controlled to stamp surface quality etc. to the sawn timber according to the surface quality grades
Grade label;
Step 410:The code-spraying equipment is controlled to stamp stress levels label to the sawn timber according to the stress levels.
Sawn timber surface quality grades disclosed by the invention divide device and method, and sawn timber is analyzed using vision inspection apparatus 7
Surface whether meet preset condition, so as to using machine vision technique carry out range estimation classification to sawn timber, improve sawn timber range estimation
The accuracy of hierarchical detection and efficiency.Meanwhile using the volume and type of image analysis sawn timber, and identify sawn timber annual ring, utilize
Weighing sensor measures quality, with reference to sawn timber volume so as to obtain the density of sawn timber, according to density, the type of sawn timber, annual ring
Information, moisture content, surface defect, with reference in sawn timber density in experimental data base in control machine, the type of sawn timber, annual ring information, contain
Elasticity modulus corresponding to water rate, surface defect gives tested sawn timber one preliminary stress grading, is without mechanical stress point
The sawn timber of grade provides preliminary stress grading reference, moreover, the present invention realizes multiple device continuous actions, realizes lumber class
The automation of division improves the efficiency of lumber class division.
Specific case used herein is expounded the principle of the present invention and embodiment, and above example is said
The bright method and its core concept for being merely used to help understand the present invention;Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, in specific embodiments and applications there will be changes.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.