CN106403853A - Stone material surface roughness on-line detection device and stone material surface roughness on-line detection method - Google Patents

Abstract

The invention puts forward a stone material surface roughness on-line detection device and a stone material surface roughness on-line detection method, and belongs to the technical field of stone material surface quality detection. The device comprises a base, a liftable bracket, a camera, an enlarger lens, a circular light source, a camera obscura, a positioning device, an image acquisition card, and a computer. The method has a simple process, and is easy to operate. Problems such as low precision and unstable detection caused by subjective factors and proficiency are overcome. An optical technology, a computer technology, a digital image processing technology and an on-line precision measurement technology are integrated. The surface glossiness of a stone material can be detected online, the detection result can be fed back to a detection system in real time, and stone roughness detection can be completed automatically. A plurality of problems of stone enterprises, such as low detection efficiency and unstable detection, are solved effectively.

Description

A kind of on-line measuring device of stone surface roughness and method

Technical field

The invention belongs to stone surface technical field of quality detection is and in particular to a kind of online inspection of stone surface roughness Survey device and method.

Background technology

In recent years, what the development with domestic construction industry and people's living standard and low-carbon environment-friendly were realized constantly carries Height, has coloured and dazzling lithotome and enjoys in building decoration and like, yet with China's stone material production technology level Limit, lead to product quality uneven, therefore stone surface detection is particularly important.

At present, China's overwhelming majority enterprise still adopts traditional manual detection method, and stone material is placed on artificial vision's model In enclosing, adjust the irradiating angle of light first, then by observing the detection that product is carried out with surface quality, finally according to detection Result is carrying out rank differentiation.The detection experience of quality inspection personnel thus product quality and operating efficiency place one's entire reliance upon, by subjectivity The factor and impact of qualification factor is larger requires it is difficult to meet high accuracy, high efficiency and stable detection etc., additionally, in addition people The Operating Complexity of work detection, is therefore extremely necessary to invent a kind of new online test method, thus making up prior art Present in not enough.

Content of the invention

For the deficiencies in the prior art, the present invention proposes a kind of on-line measuring device of stone surface roughness and method, To solve present in prior art, to need manual operations, complex operation with this, precision is low, efficiency is low and does not reach stable detection The problems such as requirement.

A kind of on-line measuring device of stone surface roughness, including base, liftable support, video camera, enlarging lens, Annular light source, camera bellows, positioner, image pick-up card and computer；

Wherein, described base side is fixedly installed liftable support, and the measurement head of liftable support is fixedly installed Video camera, the camera lens lower end of video camera is provided with enlarging lens；Described base upper ends have positioner, on positioner Slabstone to be measured is placed at end, is covered with camera bellows outside positioner；Described video camera stretches into camera bellows through camera bellows top, described camera bellows Annular light source is fixedly installed on the wall of side, and the camera lens of video camera is located at the center of annular light source；Described described video camera Output end connects the input of image pick-up card, and the output end of image pick-up card connects the input of computer.

Described base, its lower end is provided with isolation cushion.

Described liftable support, including：Column cylinder, bearing, ball-screw, ball guide screw nat, measurement head, outstanding Arm, shaft coupling and the first stepper motor, wherein, described column cylinder is fixedly installed on the side of base, and column cylinder passes through Bearing connects ball-screw, and leading screw connects stepper motor by shaft coupling, and described ball-screw is additionally provided with ball-screw spiral shell Female secondary, described ball guide screw nat is fixedly connected with one end of cantilever, and the other end of cantilever is fixedly connected measurement head.

Described video camera adopts ccd video camera, and described enlarging lens adopt 600 times of enlarging lens.

Described positioner, including the objective table with slide block, slide rail, guide rail platform, ball-screw and the second stepping electricity Machine；Wherein, described guide rail platform is fixed on base, and the second described stepper motor is fixed on the side of guide rail platform, rolling Ballscrew is fixed on guide rail platform upper end, and the output shaft of the second stepper motor connects one end, ball-screw and the load of ball-screw Thing platform is threaded, and the slide block of objective table lower end is along the slide rail slip of guide rail platform upper end, objective table upper ends stone to be measured Plate.

The detection method that on-line measuring device according to stone surface roughness is carried out, comprises the following steps：

Step 1, the upper end of positioner stone material to be measured being positioned in camera bellows；

Step 2, regulation liftable support make video camera be in required height, open annular light source；

Step 3, the lens location of adjustment video camera, are made focusing surface on the plane of reference of stone material to be measured, are led to using video camera Cross enlarging lens and image is taken to the clear position bat of some point of slabstone；

Step 4, regulation liftable support make video camera rise according to the spacing distance setting, and stone plate surface are carried out continuous Shoot, till clapping the picture blur taking, and then the Series microscopy images of acquisition complete set, and sent out by image pick-up card Deliver in computer, to be captured complete after, step motor control camera lens in positioner return initial position；

Step 5, start positioner in stepper motor drive ball-screw motion, make objective table laterally walk setting away from From, and then mobile slabstone to be measured, repeat step 3 to step 5, to stone plate surface, continuously multiple points shoot, thus obtaining many Cover complete Series microscopy images, and sent to computer by image pick-up card；

Step 6, using the self-adaptive smooth filtering algorithm being improved based on traditional gaussian filtering method to sequence chart As being filtered processing, carry out the blending algorithm based on color notation conversion space after completing image registration and carry out image co-registration；

Step 7, using Measure Algorithm obtain multiple points image depth information；

Step 8, recover the exact depth information of the multiple point of stone plate surface to be measured by interpolation fitting, and then construct and treat Survey the two-dimensional transversal profile of stone plate surface roughness, and using this two-dimensional curve as object, obtain the mean roughness of multiple points Value, and use it to represent the surface roughness of tested slabstone.

Described in step 6 using the self-adaptive smooth filtering algorithm being improved based on traditional gaussian filtering method to sequence Row image is filtered processing, specific as follows：

Step 6-1, setting maximum iteration time, arrange step-length；

Step 6-2, acquisition sequence image pixel (x, y) are in the gradient of X-direction and gradient in the Y direction；

Step 6-3, according to each pixel in the gradient of X-direction and gradient in the Y direction, obtain wave filter weight coefficient；

Wherein, w^(k)(x, y) represents pixel (x, y) wave filter weight coefficient in kth time iteration, G_x ^(k)(x, y) represents kth In secondary iteration, pixel is in the gradient of X-direction；G_x ^(k)(x, y) represents pixel gradient in the Y direction in kth time iteration；H represents Step-length；

Step 6-4, by weighted average obtain image slices vegetarian refreshments gray value；

Wherein, f^(k+1)(x, y) represents the gray value of pixel (x, y) in+1 iteration of kth, f^(k)(x+i, y+j) represents the The gray value of image slices vegetarian refreshments (x+i, y+j), w in k iteration^(k)(x+i, y+j) represents pixel (x+i, y+ in kth time iteration J) wave filter weight coefficient,

Step 6-5, judge whether iteration reaches maximum iteration time, if so, then stop, completing to filter, otherwise, return is held Row step 6-2.

Average roughness value computing formula described in step 8 is specific as follows：

Wherein, Ra represents average roughness value, and n is IMAQ point number, y_iProfile offset distance for i-th point.

Advantage of the present invention：

With vision technique as development platform, design a kind of stone surface roughness on-line measuring device and method, flow process It is simple and convenient to operate；Overcome because the problems such as the precision caused by subjective factor and qualification factor is low, detection is unstable； Integrate Application Optics technology, computer technology, Digital image processing technique, online Technology of Precision Measurement, can examine online Survey stone surface glossiness, by testing result Real-time Feedback to detecting system, be automatically performed the detection of stone material roughness, effectively Solve the problems such as stone material enterprise detection amount efficiency is low, detection is unstable.

Brief description

Fig. 1 is the on-line measuring device structural representation of the stone surface roughness of an embodiment of the present invention, wherein, 1 Represent base, 2 expression liftable supports, 3 expression video cameras, 4 expression enlarging lens, 5 expression annular light sources, 6 expression camera bellows, 7 Represent stone material to be measured, 8 expression positioners, 9 expression image pick-up cards, 10 expression computers, 11 expression isolation cushions；

Fig. 2 is the liftable support schematic diagram of an embodiment of the present invention, and wherein, 2-1 represents that column cylinder, 2-2 represent Bearing, 2-3 represent that ball-screw, 2-4 represent that ball guide screw nat, 2-5 represent that measurement head, 2-6 represent that cantilever, 2-7 represent Shaft coupling, 2-8 represent the first stepper motor；

Fig. 3 is the positioner schematic diagram of the present invention, and wherein, 8-1 represents the objective table with slide block, and 8-2 represents slide rail, 8- 3 expression guide rail platforms, 8-4 represents screw, and 8-5 represents ball-screw, and 8-6 represents the second stepper motor；

Fig. 4 is the online test method flow chart of the stone surface roughness of an embodiment of the present invention.

Specific embodiment

Below in conjunction with the accompanying drawings an embodiment of the present invention is described further.

In the embodiment of the present invention, as shown in figure 1, the on-line measuring device of stone surface roughness includes base 1, liftable Support 2, video camera 3, enlarging lens 4, annular light source 5, camera bellows 6, slabstone to be measured 7, positioner 8, image pick-up card 9, calculating Machine 10, isolation cushion 11；

In the embodiment of the present invention, video camera 3 adopts CMV-500UC industrial digital ccd video camera, described enlarging lens 4 Using 600 times of enlarging lens, ccd video camera 3 and enlarging lens 4 assemble as imageing sensor；

In the embodiment of the present invention, amplifier is assembled with CCD industry shooting, by adjusting field-of-view angle and width, To obtain part to be collected clearly surface information.

In the embodiment of the present invention, described base 1 side is fixedly installed liftable support 2, the measurement of liftable support 2 Video camera 3 is fixed, the camera lens lower end of video camera 3 is provided with enlarging lens 4 on head；Described base 1 upper end stationary positioned device 8, positioner upper ends slabstone to be measured 7, it is covered with camera bellows 6 outside fixing device；Described video camera 3 passes through camera bellows 6 top Stretch into camera bellows 6, described camera bellows 6 one the week side of boss wall is fixedly installed annular light source 5, and the camera lens of video camera 3 is located at annular light source 5 Center；Described base 1, its lower end is provided with isolation cushion 11；The output end of described video camera 3 connects image pick-up card 9 Input, the output end of image pick-up card 9 connects the input of computer 10.

In the embodiment of the present invention, as shown in Fig. 2 described liftable support 2 includes：Column cylinder 2-1, bearing 2-2, rolling Ballscrew 2-3, ball guide screw nat 2-4, measurement head 2-5, cantilever 2-6, shaft coupling 2-7 and the first stepper motor 2-8；Its In, described column cylinder 2-1 is fixedly installed on the side of base 1, and column cylinder 2-1 connects ball-screw by bearing 2-2 2-3, leading screw connects the first stepper motor 2-8 by shaft coupling 2-7, and described ball-screw 2-3 is additionally provided with ball-screw spiral shell Female secondary 2-4, described ball guide screw nat 2-4 are fixedly connected with one end of cantilever 2-6, and the other end of cantilever 2-6 is fixedly connected Measurement head 2-5；

In the embodiment of the present invention, liftable support 2 effect is for installing ccd video camera and controlling its rising, by stepping The mode that motor is combined with ball-screw to guarantee that ccd video camera can move along vertical tested stone slab surface Z axis；Vertical Post cylinder and axle mainly play a supportive role；Shaft coupling is used for fixing axle；The Main Function of ball-screw is to change rotary motion For linear movement, there is the features such as high accuracy and high efficiency simultaneously, and there is accurate Fine Feed function；Therefore in column The upper method combining with ball-screw with stepper motor can realize high accuracy, efficient linear movement；Ball-screw Pair of nut is used for fixing leading screw；Cantilever is used for connecting measurement head, rises and connects and supporting role；Measurement head refers to the magnifying glass assembling With CCD industrial camera, for measuring.

In the embodiment of the present invention, as shown in figure 3, described positioner 8 includes：With the objective table 8-1 of slide block, slide rail 8- 2, guide rail platform 8-3, screw 8-4, ball-screw 8-5, the second stepper motor 8-6；Wherein, described guide rail platform 8-3 passes through Screw 8-4 is fixed on base 1, and the second described stepper motor 8-6 is fixed on the side of guide rail platform 8-3, ball-screw 8-5 It is fixed on guide rail platform 8-3 upper end, the output shaft of the second stepper motor 8-6 connects one end of ball-screw 8-5, ball-screw 8- 5 are threadeded with objective table 8-1, and the slide block of objective table 8-1 lower end is along the slide rail 8-2 slip of guide rail platform 8-3 upper end, loading Platform 8-1 upper ends slabstone to be measured；

In the embodiment of the present invention, described positioner 8 is for transverse shifting slabstone to be measured, and then shoots stone plate surface The sequence image of continuous 5 points；Guide rail platform mainly plays a supportive role, and screw plays fixing guide rail platform action, and guide rail is used for carrying The slip of the objective table of slide block, steady pin is used for fixing ball-screw, and stepper motor transmits power, and ball-screw is by rotary motion Be converted to linear motion, the objective table with slide block carries slabstone transverse movement to be measured.

In the embodiment of the present invention, the detection method that carried out using the on-line measuring device of stone surface roughness, method stream Journey figure is as shown in figure 4, comprise the following steps：

Step 1, positioner upper end stone material to be measured being positioned in camera bellows；

Step 2, regulation liftable support make video camera be in required height, open annular light source；

In the embodiment of the present invention, adjustment lifting platform ensures that the first stepper motor is in required height, and next is opened and adjusts Section LED annular light source, to ensure sufficient brightness in camera bellows, illuminates stone material plate surface, in order to IMAQ；

Step 3, the lens location of adjustment video camera, are made focusing surface on the plane of reference of stone material to be measured, are led to using video camera Cross enlarging lens and image is taken to the clear position bat of some point of slabstone；

Step 4, regulation liftable support make video camera rise according to the spacing distance setting, and stone plate surface are carried out continuous Shoot, till clapping the picture blur taking, and then the Series microscopy images of acquisition complete set, and sent out by image pick-up card Deliver in computer, to be captured complete after, step motor control camera lens in positioner return initial position；

In the embodiment of the present invention, gather piece image every 52 μm；

Step 5, the stepper motor starting in positioner drive ball-screw motion, make objective table laterally walking necessarily away from From, and then mobile slabstone to be measured, repeat step 3,4 and 5,5 points continuous to stone plate surface shoot, thus obtain 5 sets complete Whole Series microscopy images, and sent to computer by image pick-up card.

In the embodiment of the present invention, subsequently control camera lens to return initial position, so that the Series microscopy images of collection is comprised tested The information of the whole Z-direction of large stone material plate, each image has focus and fuzzy region, then logical in sequence image Cross certain fusion rule and carry out image co-registration and obtain each pixel corresponding focus position, and then reconstruct 5 width not Same total focus image；

Step 6, using the self-adaptive smooth filtering algorithm being improved based on traditional gaussian filtering method to sequence chart As being filtered processing, carry out the blending algorithm based on color notation conversion space after completing image registration and carry out image co-registration；

In the embodiment of the present invention, described is filtered using the adaptive smooth being improved based on traditional gaussian filtering method Ripple algorithm is filtered to sequence image processing, specific as follows：

Step 6-1, setting maximum iteration time, arrange step-length；

Step 6-2, acquisition sequence image pixel (x, y) are in the gradient of X-direction and gradient in the Y direction；

Step 6-3, according to each pixel in the gradient of X-direction and gradient in the Y direction, obtain wave filter weight coefficient；

Wherein, w^(k)(x, y) represents pixel (x, y) wave filter weight coefficient in kth time iteration, G_x ^(k)(x, y) represents kth In secondary iteration, pixel is in the gradient of X-direction；G_x ^(k)(x, y) represents pixel gradient in the Y direction in kth time iteration；H represents Step-length；

Step 6-4, by weighted average obtain image slices vegetarian refreshments gray value；

Wherein, f^(k+1)(x, y) represents the gray value of pixel (x, y) in+1 iteration of kth, f^(k)(x+i, y+j) represents the The gray value of image slices vegetarian refreshments (x+i, y+j), w in k iteration^(k)(x+i, y+j) represents pixel (x+i, y+ in kth time iteration J) wave filter weight coefficient,

Step 6-5, judge whether iteration reaches maximum iteration time, if so, then stop, completing to filter, otherwise, return is held Row step 6-2；

Step 7, using Measure Algorithm obtain continuous 5 points image depth information；

In the embodiment of the present invention, after completing image co-registration, calculate image depth information, Focus field emission array using Measure Algorithm The principle of algorithm is that selective focus operator, selects variance focusing operator in the embodiment of the present invention, and with FM_VCan serve as gathering Jiao estimates operator to recover figure depth information；

Wherein, FM_VRepresent variance Focus field emission array operator；N represents iterations；P (x, y) represents p pixel point coordinates；G (x, Y) represent the gray scale coordinate of g pixel；U(x₀, y₀) represent neighborhood of pixel points；Represent neighborhood U (x₀, y₀) gray scale Mean value；S represents template；M represents the pixel total number comprising current pixel in this template；F (x, y) represents original image Gray value；

Step 8, recover the more accurate depth information of 5 points of stone plate surface to be measured by interpolation fitting, and then construct Go out the two-dimensional transversal profile of stone plate surface roughness to be measured, and using this two-dimensional curve as calculating object, then utilize formula meter Calculate the average roughness value of 5 points, and use it to represent the surface roughness of tested slabstone.

In the embodiment of the present invention, the average roughness value computing formula described in step 8 is specific as follows：

Wherein Ra is average roughness value, and n is IMAQ point number, y_iProfile offset distance for i-th point.

In the embodiment of the present invention, recover more accurate body surface half-tone information finally by interpolation fitting, realize By two-dimensional sequence Image Reconstruction tested stone slab surface three-D profile, and it is analyzed calculate, draws specifically average Roughness value, then the numerical value measured with contact pin type contourgraph is compared analysis, finally shows in the display.

Claims (8)

1. a kind of on-line measuring device of stone surface roughness it is characterised in that include base, liftable support, video camera, Enlarging lens, annular light source, camera bellows, positioner, image pick-up card and computer；

Wherein, described base side is fixedly installed liftable support, and the measurement head of liftable support is fixedly installed shooting Machine, the camera lens lower end of video camera is provided with enlarging lens；Described base upper ends have positioner, and positioner upper end is put Put slabstone to be measured, outside positioner, be covered with camera bellows；Described video camera stretches into camera bellows through camera bellows top, described camera bellows side wall On be fixedly installed annular light source, and the camera lens of video camera is located at the center of annular light source；The output of described described video camera End connects the input of image pick-up card, and the output end of image pick-up card connects the input of computer.

2. the on-line measuring device of stone surface roughness according to claim 1 is it is characterised in that described base, Its lower end is provided with isolation cushion.

3. the on-line measuring device of stone surface roughness according to claim 1 is it is characterised in that described liftable Support, including：Column cylinder, bearing, ball-screw, ball guide screw nat, measurement head, cantilever, shaft coupling and the first stepping Motor, wherein, described column cylinder is fixedly installed on the side of base, and column cylinder connects ball-screw, silk by bearing Thick stick connects stepper motor by shaft coupling, and described ball-screw is additionally provided with ball guide screw nat, described ball-screw spiral shell Female pair is fixedly connected with one end of cantilever, and the other end of cantilever is fixedly connected measurement head.

4. the on-line measuring device of stone surface roughness according to claim 1 is it is characterised in that described video camera Using ccd video camera, described enlarging lens adopt 600 times of enlarging lens.

5. the on-line measuring device of stone surface roughness according to claim 1 is it is characterised in that described positioning fills Put, including the objective table with slide block, slide rail, guide rail platform, ball-screw and the second stepper motor；Wherein, described guide rail is put down Platform is fixed on base, and the second described stepper motor is fixed on the side of guide rail platform, and ball-screw is fixed on guide rail platform Upper end, the output shaft of the second stepper motor connects one end of ball-screw, and ball-screw is threadeded with objective table, and objective table The slide block of lower end slides along the slide rail of guide rail platform upper end, objective table upper ends slabstone to be measured.

6. the detection method being carried out using the on-line measuring device of the stone surface roughness described in claim 1, its feature exists In comprising the following steps：

Step 1, the upper end of positioner stone material to be measured being positioned in camera bellows；

Step 2, regulation liftable support make video camera be in required height, open annular light source；

Step 3, the lens location of adjustment video camera, make focusing surface on the plane of reference of stone material to be measured, pass through to put using video camera Macro lens is clapped to the clear position of some point of slabstone and is taken image；

Step 4, regulation liftable support make video camera rise according to the spacing distance setting, and stone plate surface is carried out with continuous bat Take the photograph, till clapping the picture blur taking, and then the Series microscopy images of acquisition complete set, and sent by image pick-up card To computer, to be captured complete after, step motor control camera lens in positioner return initial position；

Step 5, the stepper motor starting in positioner drive ball-screw motion, so that objective table is laterally walked setpoint distance, And then move slabstone to be measured, repeat step 3 to step 5, to stone plate surface, continuously multiple points shoot, thus obtaining covered more Whole Series microscopy images, and sent to computer by image pick-up card；

Step 6, using based on the self-adaptive smooth filtering algorithm that traditional gaussian filtering method improves, sequence image is entered Row filtering process, carries out the blending algorithm based on color notation conversion space after completing image registration and carries out image co-registration；

Step 7, using Measure Algorithm obtain multiple points image depth information；

Step 8, recover the exact depth information of the multiple point of stone plate surface to be measured by interpolation fitting, and then construct stone to be measured The two-dimensional transversal profile of plate surface roughness, and using this two-dimensional curve as object, obtain the average roughness value of multiple points, and Use it to represent the surface roughness of tested slabstone.

7. detection method according to claim 6 is it is characterised in that the employing described in step 6 is based on traditional Gauss The self-adaptive smooth filtering algorithm that filtering method improves is filtered to sequence image processing, specific as follows：

Step 6-1, setting maximum iteration time, arrange step-length；

Step 6-2, acquisition sequence image pixel (x, y) are in the gradient of X-direction and gradient in the Y direction；

Step 6-3, according to each pixel in the gradient of X-direction and gradient in the Y direction, obtain wave filter weight coefficient；

$w^{\left(k\right)}(x,y)=\exp \[-\frac{{\[{G_x}^{\left(k\right)}(x,y)\]}^2+{\[{G_x}^{\left(k\right)}(x,y)\]}^2}{2h^2}\]---\left(1\right)$

Wherein, w^(k)(x, y) represents pixel (x, y) wave filter weight coefficient in kth time iteration, G_x ^(k)(x, y) represents kth time repeatedly In generation, pixel is in the gradient of X-direction；G_x ^(k)(x, y) represents pixel gradient in the Y direction in kth time iteration；H represents step Long；

Step 6-4, by weighted average obtain image slices vegetarian refreshments gray value；

$f^{(k+1)}(x,y)=\frac{1}{N^{\left(k\right)}}\underset{i=-1}{\overset{1}{\Σ}}\underset{j=-1}{\overset{1}{\Σ}}{f}^{\left(k\right)}(x+i,y+j)w^{\left(k\right)}(x+i,y+j)---\left(2\right)$

Wherein, f^(k+1)(x, y) represents the gray value of pixel (x, y) in+1 iteration of kth, f^(k)(x+i, y+j) represents kth time The gray value of image slices vegetarian refreshments (x+i, y+j), w in iteration^(k)(x+i, y+j) represents pixel (x+i, y+j) in kth time iteration Wave filter weight coefficient,

Step 6-5, judge whether iteration reaches maximum iteration time, if so, then stop, completing to filter, otherwise, return and execute step Rapid 6-2.

8. detection method according to claim 6 is it is characterised in that the average roughness value described in step 8 calculates public affairs Formula is specific as follows：

$Ra=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}\left|yi\right|---\left(3\right)$

Wherein, Ra represents average roughness value, and n is IMAQ point number, and yi is i-th point of profile offset distance.