Disclosure of Invention
The invention aims to provide a sawn timber surface quality grade dividing device and method, which improve grading accuracy.
In order to achieve the above object, the present invention provides the following solutions:
a sawn timber surface quality grading device, comprising: the device comprises a water content detection device, a weighing sensor, a visual detection device and code spraying equipment, wherein the water content detection device, the weighing sensor, the visual detection device and the code spraying equipment are sequentially arranged along the conveying direction of sawn timber; the output ends of the water content detection device, the weighing sensor and the visual detection device are all electrically connected with the central control unit; the water content detection device is used for detecting the water content of the sawn timber; the weighing sensor is used for detecting the weight of the sawn timber; the visual detection device is used for detecting surface defects of the sawn timber and identifying annual rings of the sawn timber; the code spraying equipment is used for printing surface quality grade and primarily-estimated stress grade marks on the sawn timber under the control of the central control computer;
the visual detection device comprises six right-shot cameras, and the six cameras are respectively arranged in six directions of right above, right below, right left side, right side, right front and right rear of the position to be shot.
Optionally, the visual detection device further includes a plurality of oblique photographing cameras, and the oblique photographing cameras are located obliquely above and obliquely below the position to be photographed; the included angle between the shooting angle of each oblique shooting camera and the horizontal plane is adjustable between 20 degrees and 70 degrees.
Optionally, the level dividing device further includes a first photosensor, a second photosensor, a third photosensor, and a fourth photosensor;
the first photoelectric sensor is positioned between the starting position of the conveyor belt and the water content detection device and is close to the water content detection device; the second photoelectric sensor is positioned between the water content detection device and the weighing sensor and is close to the weighing sensor; the third photoelectric sensor is positioned between the weighing sensor and the visual detection device and is close to the visual detection device; the fourth photoelectric sensor is located between the visual detection device and the code spraying equipment and is close to the code spraying equipment.
Optionally, the visual detection device is arranged in the image acquisition room.
Optionally, an illumination device is further arranged in the image acquisition chamber.
Optionally, the level dividing device further comprises a motor; the motor is used for driving the conveying belt to drive the sawn timber to convey.
The invention also discloses a sawn timber surface quality grade dividing method which is applied to the sawn timber surface quality grade dividing device; the method comprises the following steps:
acquiring the water content detected by the water content detection device;
acquiring the weight detected by the weighing sensor;
acquiring an image shot by the visual detection device;
identifying the type, the outline, the annual rings and the surface defects of the sawn timber according to the image;
calculating the volume of the sawn timber according to the profile;
calculating the density of the sawn timber according to the volume and the weight;
dividing the quality grade of the sawn timber surface according to the surface defects;
comparing the density, the type, the annual rings of sawn timber, the surface defects and the water content with data in a database, and evaluating the stress level of the sawn timber;
controlling the code spraying equipment to print surface quality grade marks on the sawn timber according to the surface quality grade;
and controlling the code spraying equipment to print stress grade marks on the sawn timber according to the stress grade.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: according to the sawn timber surface quality grade classifying device and method disclosed by the invention, whether the surface of sawn timber meets the preset condition is analyzed by utilizing the visual detection device, so that the sawn timber is subjected to visual classification by utilizing the machine visual technology, and the accuracy and the efficiency of the visual classification detection of sawn timber are improved. Meanwhile, the volume and the type of the sawn timber are analyzed by utilizing the image, the annual rings of the sawn timber are identified, the quality is measured by utilizing the weighing sensor, the volume of the sawn timber is combined to obtain the density of the sawn timber, and the elastic modulus corresponding to the density of the sawn timber, the type of the sawn timber, the annual ring information, the moisture content and the surface defect in an experimental database in the central control machine is combined to provide a preliminary stress classification reference for the sawn timber which is not subjected to mechanical stress classification. In addition, the invention realizes the continuous action of a plurality of devices, realizes the automation of sawn timber grading and improves the sawn timber grading efficiency.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Fig. 1 is a block diagram of an apparatus for classifying saw blade surface quality according to a first embodiment of the present invention.
Referring to fig. 1, the sawn timber surface quality grading device comprises: a bench 1, a conveyor belt 2 and a motor 3. The conveying belt 2 is arranged on the rack 1, and the motor 3 is used for driving the conveying belt 2 to convey and driving the sawn timber 4 to convey.
A water content detection device 5, a weighing sensor 6, a visual detection device 7 and a code spraying device 8 are sequentially arranged along the conveying direction of the sawn timber 4; the output ends of the water content detection device 5, the weighing sensor 6 and the visual detection device 7 are electrically connected with a central control unit; the water content detection device 5 is used for detecting the water content of the sawn timber 4; the weighing sensor 6 is used for detecting the weight of the sawn timber 4; the visual detection device 7 is used for detecting surface defects of the sawn timber 4 and identifying annual rings of the sawn timber 4; the code spraying equipment 8 is used for printing grade marks on the sawn timber 4 under the control of the central control machine; the visual detection device 7 is arranged in the image acquisition room 9; an illumination device 10 is also arranged in the image acquisition chamber 9.
The level dividing means further comprises a first photosensor 11, a second photosensor 12, a third photosensor 13 and a third photosensor 14; the first photoelectric sensor 11 is located between the starting position of the conveyor belt and the water content detection device 5 and is close to the water content detection device 5, and is used for detecting whether the sawn timber 4 enters a water content detection area; the second photoelectric sensor 12 is located between the moisture content detection device 5 and the weighing sensor 6 and is close to the weighing sensor 6, and is used for detecting whether the sawn timber 4 enters a weighing area; the third photoelectric sensor 13 is located between the weighing sensor 6 and the visual detection device 7 and is close to the visual detection device 7, and is used for detecting whether the sawn timber 4 enters a visual detection area; the third photoelectric sensor 14 is located between the visual detection device 7 and the code spraying device 8 and near the code spraying device 8, and is used for detecting whether the sawn timber 4 enters a marking area. The first, second, third and third photosensors 11, 12, 13 and 14 provide start signals for the water content detection device 5, the weighing sensor 6, the visual detection device 7 and the code spraying apparatus 8, respectively.
Fig. 2 is a device structure diagram of a visual inspection device 7 according to a first embodiment of the sawn timber surface quality grade separation device of the present invention.
Referring to fig. 2, the visual inspection apparatus 7 includes six right-shot cameras, and the six cameras are respectively disposed in six directions right above, right below, right left side, right front, and right rear of the position to be shot.
The six positive shooting cameras are mainly used for detecting defects of middle sections, cluster sections, combined sections, section holes, small worm holes, big worm holes, machining offset, leakage plane, splitting, decay, oblique textures and the like of the sawn timber 4. These defects basically occur on the wide and narrow faces of the sawn timber and the end parts of the sawn timber, and only four cameras on the wide and narrow faces and two cameras on the two ends are needed for image acquisition.
Fig. 3 is a device structure diagram of a visual inspection device 7 according to a second embodiment of the sawn timber surface quality grade separation device of the present invention.
Referring to fig. 3, the visual detection device 7 further includes a plurality of oblique photographing cameras, and a plurality of oblique photographing cameras are located obliquely above and obliquely below the position to be photographed; the included angle between the shooting angle of each oblique shooting camera and the horizontal plane is adjustable between 20 degrees and 70 degrees.
The oblique photographing camera is mainly used for detecting defects such as processing damage holes, saw cuts, blunt edges, cracks, three-surface sections, strip-shaped sections, round edge sections, blue changes and the like of sawn materials. These defects occur mainly at the corners of the sides of the sawn timber and across both the width and the width of the sawn timber.
The invention can realize the identification of the saw material types. The sawn timber can naturally form natural textures in the growth process, the natural textures show certain regularity, some of the natural textures show longitudinal stripe textures, and some of the natural textures show transverse pattern textures. The contrast of the sawn timber texture is deep and shallow, and the deeper texture and the shallower texture can be detected well by the invention; different sawages have different twills, fractal geometric analysis is utilized to discuss fractal characteristics of radial and chord textures of the sawages, and the fractal dimension values of the sawages are used for representing the shape, distribution density, uniformity and width characteristics of the sawages surface textures, so that the detected sawages are judged.
The invention can realize defect type identification, and the specific process is as follows:
adopting a neural network algorithm to identify and study the types of saw material surface defects: the sawn timber surface defects have the characteristics of gray level difference, shape difference and geometric difference, the gray level characteristics, the shape characteristics and the geometric characteristics are selected to establish a characteristic database, the characteristic database is used as a judging basis of defect types, and the process of establishing the database is as follows:
the basic shape of the defect is characterized by adopting a minimum circumscribed rectangle: distinguishing slender defects (bar-shaped knots, basic blunt edges and cracks) according to the aspect ratio value of the minimum circumscribed rectangle as a characteristic value; dividing three defects of saw cuts, worm holes and pitch holes according to the rectangle degree value;
the position information of the defects is reflected by establishing a coordinate system, and the method can be used for defects such as middle joints, clustered joints, combined joints, joint holes, small worm holes, big worm holes, machining offset, leakage plane, splitting, decay, oblique textures and the like. These defects occur substantially on both the wide and narrow faces of the log, the abscissa (the co-ordinate of the log in the axial direction) being very effective for distinguishing these defects, while the ordinate (the co-ordinate of the log in the radial direction) is not very effective for distinguishing the types of defects, and need not be employed.
The average gray value and the gray mean square error reflect the gray characteristics of the defects, can reflect the bright and dark characteristics of the defects, have obvious differences in the average gray value and the gray mean square error of the dead knots, the movable knots and the sawn timber with blue parts, and after the database is built, the minimum circumscribed rectangle of the defects, the area ratio of the areas of the defects to the minimum circumscribed rectangle, the abscissa of the defects, the average gray value and the average gray mean square error of the defects are determined by analyzing and processing the acquired images, and the defects to be identified are identified by comprehensively considering the factors, wherein the specific process is as follows:
according to the invention, the extracted characteristics are processed, 7 characteristic values are selected as classification indexes of the neural network for saw material surface defects, wherein the classification indexes comprise 7 indexes of width-to-length ratio, direction offset, full-length ratio of abscissa, logarithm of rectangle degree and density, gray level ratio of average gray level value and gray mean square error, and the indexes are used for judging saw material surface defects. And taking 19 defects to be identified as output of the neural network to the judging result.
And obtaining the type of the defect according to the calculation of the system to the seven characteristic values and the comparison analysis of the seven characteristic values and the expected value.
The invention can realize the identification of the annual rings of sawn timber, and specifically comprises the following steps: aiming at the characteristics of tree annual ring images, the data of the annual ring images are compressed by utilizing a wavelet transformation method, a higher compression rate is obtained on the premise of ensuring higher definition, then the annual ring images are segmented and analyzed, and finally the sawn timber annual rings are identified.
The invention can also judge whether the knots on the surface of the sawn timber meet preset conditions, and the specific process is as follows: acquiring an image of the sawn timber surface; identifying knots on the surface of the sawn timber by utilizing an image identification technology; judging whether the positions, the sizes and the number of the knots accord with preset values or not, and determining that the sawn timber is a disqualified sawn timber when any one of the positions, the sizes and the number of the knots does not accord with the preset values.
The invention can also judge whether the late wood rate of the sawn timber meets the preset condition of the late wood rate, and the specific process is as follows:
identifying a growing wheel on the surface of the sawn timber and an edge area of the late timber in each growing wheel by utilizing texture identification;
calculating the width of the growth wheel and the width of the late material in each growth wheel;
the late wood rate was calculated using the following formula:
wherein: l (L) w The late-cut rate of the test sample,
ΣL b measuring the total width of the evening material in mm
b-total annual wheel width in mm in the measurement range
Comparing the late wood rate with a preset late wood rate threshold, and determining that the late wood rate does not meet a preset condition when the late wood rate exceeds the preset late wood rate threshold, and determining that the sawn wood is unqualified sawn wood; and when the late wood rate meets the preset late wood rate threshold, determining that the late wood rate meets a preset condition, and determining that the sawn timber is qualified sawn timber.
The invention can also judge whether the blue change proportion of the sawn timber meets the preset blue change condition, and the specific process is as follows:
identifying a region of the sawn timber surface where blue change occurs;
calculating the area of the area with blue change and the total area of sawn timber;
calculating the blue transformation ratio according to the area of the blue transformation-generating area and the total area of the sawn timber;
comparing the blue transformation ratio with a preset blue transformation ratio threshold, and determining that the sawn timber is an unqualified sawn timber when the blue transformation ratio exceeds the preset blue transformation ratio threshold; and when the blue change proportion meets the preset blue change proportion threshold, determining that the sawn timber is qualified sawn timber.
Fig. 4 is a flow chart of a method for classifying saw material surface quality according to an embodiment of the present invention.
Referring to fig. 4, the sawn timber surface quality grading method is applied to the sawn timber surface quality grading device; the method comprises the following steps:
step 401: acquiring the water content detected by the water content detection device 5;
step 402: acquiring the weight detected by the weighing sensor 6;
step 403: acquiring an image shot by the visual detection device 7;
step 404: identifying the type, contour and surface defects of the sawn timber 4 according to the image;
step 405: calculating the volume of the sawn timber 4 according to the profile;
step 406: calculating the density of the sawn timber 4 from the volume and the weight;
step 407: dividing the surface quality grade of the sawn timber 4 according to the surface defects;
step 408: comparing the density, the type, the annual rings of sawn timber, the surface defects and the water content with data in a database, and evaluating the stress level of the sawn timber;
step 409: controlling the code spraying equipment to print surface quality grade marks on the sawn timber according to the surface quality grade;
step 410: and controlling the code spraying equipment to print stress grade marks on the sawn timber according to the stress grade.
According to the sawn timber surface quality grade classifying device and method disclosed by the invention, whether the surface of sawn timber meets the preset condition is analyzed by utilizing the visual detection device 7, so that the sawn timber is subjected to visual classification by utilizing the machine visual technology, and the accuracy and the efficiency of the visual classification detection of sawn timber are improved. Meanwhile, the volume and the type of the sawn timber are analyzed by utilizing the image, the annual ring of the sawn timber is identified, the quality is measured by utilizing the weighing sensor, the volume of the sawn timber is combined to obtain the density of the sawn timber, and the elastic modulus corresponding to the density, the type, the annual ring information, the water content and the surface defect of the sawn timber in an experimental database in the central control machine is combined to provide a preliminary stress classification for the sawn timber to be measured according to the density, the type, the annual ring information, the water content and the elastic modulus corresponding to the surface defect of the sawn timber in the central control machine, so that a preliminary stress classification reference is provided for the sawn timber which is not subjected to mechanical stress classification.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.