CN103049647B - A kind of disposal route of adding up focal plane indium post array heights data - Google Patents

Abstract

The invention discloses a kind of disposal route of adding up focal plane indium post array heights data.The present invention carries out data processing according to the features of shape of focal plane device indium post to the focal plane surface topography data obtained by laser confocal microscope, final indium post height number and the statistical distribution thereof obtaining whole device.The present invention includes: 1, obtain focal plane device surface topography data matrix, and generate corresponding coordinates matrix; 2, by calculating the base altitude determining the statistical study of indium post height; 3, utilize base altitude to be separated the altitude information matrix of single indium post; 4, calculate the height of single indium post; 5, add up the height of all indium posts.Advantage of the present invention is: 1, utilizes the present invention can realize the robotization of indium post height statistics, for the inverse bonding interconnection process of focal plane device provides foundation; 2, data processing method provided by the invention, its statistics is objective and accurate, gets rid of the data imperfection of artificial reconnaissance.

Description

A kind of disposal route of adding up focal plane indium post array heights data

Technical field:

The present invention relates to the inverse bonding field of interconnects of infrared focal plane array and sensing circuit.Achieve the robotization of focal plane indium post array heights statistics, solve the data imperfection problem that complicate statistics exists.

Background technology:

Infrared focal plane detector is day by day towards large face battle array future development, and this difficulty that inverse bonding as large face battle array focal plane device gordian technique is interconnected increases greatly, and the quality of indium post finally determine inverse bonding interconnection can complete in high quality.The detection of current indium post height generally adopts the method for microphoto observation and grab sample calculating mean value.These two class methods have certain limitation.Because indium post surface is not desirable plane, microphoto observes the concrete numerical value and distribution thereof that are difficult to obtain indium post, and grab sample is measured and can not be obtained complete indium post altitude information distribution, and easily introduces the error of artificial calibration.

The present invention carries out Treatment Analysis to the focal plane surface topography data that laser co-focusing obtains originally, specifically refers to the height distribution calculating indium post on infrared focus plane.The present invention can obtain and obtain true altitude value and the statistical distribution thereof of a large amount of indium post, has great role to the development of the inverse bonding interconnection process of infrared focal plane device.

Summary of the invention:

1, goal of the invention: the height of statistics focal plane device surface indium post and distribution thereof.

2, technical scheme:

Concrete steps of the present invention are as follows:

Step one: utilize laser confocal microscope focal plane surface to scan, obtains and preserves focal plane surface elevation topographic data matrix Z=[z _ij], wherein z _ijrefer to that coordinate is positioned at x=i, the altitude information at y=j place, the span of i, j is [1, n], and n is sampling density, value 512 or 1024;

Step 2: generate the horizontal ordinate matrix X=[x that altitude information matrix Z is corresponding _ij], ordinate matrix Y=[y _ij], wherein, matrix element meets x _ij=j, y _ij=i;

Step 3: the maximal value z obtaining element in altitude information matrix Z _max=max (z _ij), minimum value z _min=min (z _ij), generate vectorial S=[s with the interval delta z specified _h], meet s _h=z _min+ (h-1) Δ z, wherein interval delta z is set to 0.1, h round numbers 1,2 ... | (z _max-z _min)/Δ z|;

Step 4: solve level line coordinates matrix C _sh

C _sh=contour(Z,[s _hs _h]) (1)

In formula, contour () is the function in Matlab7.0, level line coordinates matrix C _shexpression is:

$C_{\mathrm{sh}}=\left[\begin{array}{cccccccc}{s}_h& \mathrm{xdata}\left(1\right)\mathrm{xdata}\left(2\right)& ...\mathrm{xdata}\left(\dim 1\right)& s_h& \mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...& \mathrm{xdata}\left(\dim 2\right)& s_h...\\ \dim 1& \mathrm{ydata}\left(1\right)\mathrm{ydata}\left(2\right)& ...\mathrm{ydata}\left(\dim 1\right)& \dim 2& \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...& \mathrm{ydata}\left(\dim 2\right)& \dim 3...\end{array}\right]---\left(2\right)$

In formula, $\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\end{array}\right]$ For the coordinate position at every bar level line place, the first behavior horizontal ordinate, the ordinate that the second behavior is corresponding, dim1, dim2, dim3 ... for the quantity that every bar level line is put, s in formula (2) is added up respectively to elements all in vectorial S _hquantity, statistics is designated as m _h, and generate level line quantity vector M=[m _h]; Generate " level line quantity-altitude information " relation curve, wherein horizontal ordinate is vectorial S, and ordinate is level line quantity vector M; In conjunction with indium cylindricality looks feature, choosing specific between first peak and second peak in " level line quantity-altitude information " relation curve is highly the base altitude ZBasic of indium post Elevation Analysis;

Step 5: the level line coordinates matrix C solving base altitude ZBasic place _zbasic

C _zbasic=contour(Z,[ZBasic ZBasic]) (3)

Level line coordinates matrix C _zbasicexpression is:

$\mathrm{Czbasic}=\left[\begin{array}{cccccccc}\mathrm{ZBasic}& \mathrm{xdata}\left(1\right)\mathrm{xdata}\left(2\right)& ...\mathrm{xdata}\left(\dim 1\right)& \mathrm{ZBasic}& \mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...& \mathrm{xdata}\left(\dim 2\right)& \mathrm{ZBasic}...\\ \dim 1& \mathrm{ydata}\left(1\right)\mathrm{ydata}\left(2\right)& ...\mathrm{ydata}\left(\dim 1\right)& \dim 2& \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...& \mathrm{ydata}\left(\dim 2\right)& \dim 3...\end{array}\right]---\left(4\right)$

With the ZBasic element value in formula for mark, isolate a series of level line coordinates matrix:

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 1\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 1\right)\end{array}\right]---\left(5\right)$

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 2\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 2\right)\end{array}\right]---\left(6\right)$

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 3\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 3\right)\end{array}\right]---\left(7\right)$

……；

Step 6: formula (5) is denoted as:

$\left[\begin{array}{c}\mathrm{Xdata}1\\ \mathrm{Ydata}1\end{array}\right]---\left(8\right)$

In formula, vectorial Xdata1, Ydata1 are:

Xdata1=[xdata(1) xdata(2) ... xdata(dim1)] (9)

Ydata1=[ydata(1) ydata(2) ... ydata(dim1)] (10)

Obtain vectorial Xdata1 respectively, in vectorial Ydata1 element maximal value and round and obtain Xdata1 _maxand Ydata1 _max, ask its minimum value respectively and round and obtain Xdata1 _minand Ydata1 _min, generate horizontal ordinate vector X _1s:

X _1s＝[xdata1 _minxdata1 _min+1xdata1 _min+2 ... xdata1 _max] (11)

Ordinate vector Y _1s:

Ｙ _１s＝[ydata1 _minydata1 _min+1ydata1 _min+2 ... ydata1 _max] (12)

Solve [X ₁, Y ₁]:

[X ₁,Y ₁]＝meshgrid(X _1s,Y _1s) (13)

Generate horizontal ordinate matrix X ₁, ordinate matrix Y ₁, in formula (13), meshgrid () is the function in Matlab7.0;

Step 7: obtain horizontal ordinate matrix X ₁, ordinate matrix Y ₁corresponding altitude information matrix Z ₁=[z1 _kl], Z ₁middle element meets z1 _kl=z _kl, wherein k, l are subscript, and have:

xdata1 _min＜k＜xdata1 _max，ydata1 _min＜l＜ydata1 _max

The method in step 4 is utilized to obtain matrix Z ₁corresponding " level line quantity-altitude information " relation curve, utilize this relation curve, in conjunction with the pattern feature on indium post surface, the approximate altitude value ZApro1 of the level line quantity chosen in curve near second peak to be the height of the assigned address of 1 be single indium post;

Step 8: the C solving single indium post approximate altitude value ZApro1 place _zapro1

C _zapro1=contour(Z ₁,[ZApro1 ZApro1]) (14)

C in formula _zapro1expression is:

$\mathrm{Czapro}1=\left[\begin{array}{ccc}{\mathrm{ZApro}}_1& \mathrm{xdata}\left(1\right)\mathrm{xdata}\left(2\right)& ...\mathrm{xdata}\left(\dim 1\right)\\ \dim 1& \mathrm{ydata}\left(1\right)\mathrm{ydata}\left(2\right)& ...\mathrm{ydata}\left(\dim 1\right)\end{array}\right]---\left(15\right)$

With the ZApro1 value in formula for mark, have and only have an isocontour coordinates matrix

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 1\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 1\right)\end{array}\right]---\left(16\right)$

Solve single indium post height height ₁:

height ₁=mean(z1 _pq) (17)

Wherein p, q is subscript, meet xdata (1) < p < xdats (dim1), ydata (1) < q < ydata (dim1);

Step 9: to formula (6), formula (7) ... a series of level line coordinates matrix respectively repeats steps six, seven, eight, draws corresponding a series of indium post height value height ₁, height ₂, height ₃, generate indium post height value vector H,

H＝[height ₁,height ₂,height ₃…] (18)

Solve hist (H), obtain the statistic histogram of indium post height, wherein hist () is the function in Matlab7.0.

The present invention has the following advantages:

1, utilize the present invention can obtain height value and the statistical distribution thereof of a large amount of indium post on focal plane, may be used for the growth quality of indium post on objective judgement focal plane, provide foundation to the inverse bonding interconnection process of infrared focal plane array and sensing circuit.

2, data processing method provided by the invention, its statistics is objective and accurate, gets rid of the data imperfection that people is reconnaissance.

Accompanying drawing illustrates:

Accompanying drawing 1: method flow diagram of the present invention.

Accompanying drawing 2: the surface topography map of the example of process of the present invention.

Accompanying drawing 3: the indium post shape appearance figure of the Generating Data File utilizing laser confocal microscope to obtain.

The level line distributed number figure that accompanying drawing 4: one complete data file generate.

Accompanying drawing 5: the indium post shape appearance figure that isolated single indium post altitude information is drawn.

The typical single indium post level line distributed number figure of 6: two, accompanying drawing.

Accompanying drawing 7: 16 complete data file are carried out to the height statistic histogram adding up all complete indium post obtained.

Embodiment:

Below in conjunction with drawings and Examples, the present invention is further described:

The data file that the present embodiment uses to be the resolution scanned by OLYMPUS-OLS3000 laser co-focusing instrument be 1024 × 1024 focal plane sample surface morphology altitude information array.Fig. 2 is shown in by the laser co-focusing microphoto of focal plane sample surface morphology, and the circular speck in Fig. 2 is indium post.The present embodiment is analyzed scan 16 data files.

First first data file is analyzed, Matlab7.0 is used to import surface topography altitude information array Z, method to specifications described in middle step one, step 2 generates X, Y-coordinate array, utilizes mesh () function to generate the surface topography map of infrared focal plane array as shown in Figure 3;

Then, middle step 3 (arranging appointed interval Δ z is 0.1um), method described in step 4 to specifications, make " level line quantity-altitude information " relation curve as shown in Figure 4, in Fig. 4, " level line quantity-altitude information " relation curve has two peaks.The present embodiment gets the base altitude that height corresponding to minimum value between two peak values is indium post Elevation Analysis;

Then, middle method described in step 5 isolates the surface topography altitude information of single indium post to specifications.Utilize the surface topography altitude information of isolated single indium post, draw the surface topography map of single indium post as shown in Figure 5 with mesh () function;

Then, method to specifications described in middle step 6, step 7, step 8, the surface topography altitude information of single indium post is analyzed, obtains " level line quantity-altitude information " relation curve of single indium post as shown in Figure 6, and finally obtain the height height of single indium post;

Finally, method to specifications described in middle step 9, obtain the height of all indium posts in first laser co-focusing data file, again above-mentioned steps is repeated to remaining 15 laser co-focusing data file, finally obtain indium post altitude informations all in 16 data files, utilize hist () function to add up all indium post height values obtained, draw statistic histogram.Fig. 7 is the height distribution statistics histogram of all indium posts that the present embodiment obtains.The indium post of this sample is highly concentrated in 8.5-9.0 μm, meets normal distribution law.

Claims (1)

1. add up a disposal route for focal plane indium post array heights data, it is characterized in that comprising the following steps:

Step one: utilize laser confocal microscope focal plane surface to scan, obtains and preserves focal plane surface elevation topographic data matrix Z=[z _ij], wherein z _ijrefer to that coordinate is positioned at x=i, the altitude information at y=j place, the span of i, j is [1, n], and n is sampling density, value 512 or 1024;

Step 2: generate the horizontal ordinate matrix X=[x that altitude information matrix Z is corresponding _ij], ordinate matrix Y=[y _ij], wherein, matrix element meets x _ij=j, y _ij=i;

Step 3: the maximal value z obtaining element in altitude information matrix Z _max=max (z _ij), minimum value z _min=min (z _ij), generate vectorial S=[s with the interval △ z specified _h], meet s _h=z _min+ (h-1) △ z, wherein △ z in interval is set to 0.1, h round numbers 1,2 ... | (z _max-z _min)/△ z|;

Step 4: solve level line coordinates matrix C _sh

C _sh＝contour(Z,[s _hs _h]) (1)

In formula, contour () is the function in Matlab7.0, level line coordinates matrix C _shexpression is:

$C_{\mathrm{sh}}=\left[\begin{array}{cccccccc}{s}_h& \mathrm{xdata}\left(1\right)\mathrm{xdata}\left(2\right)& ...\mathrm{xdata}\left(\dim 1\right)& s_h& \mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...& \mathrm{xdata}\left(\dim 2\right)& s_h...\\ \dim 1& \mathrm{ydata}\left(1\right)\mathrm{ydata}\left(2\right)& ...\mathrm{ydata}\left(\dim 1\right)& \dim 2& \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...& \mathrm{ydata}\left(\dim 2\right)& \dim 3...\end{array}\right]---\left(2\right)$

In formula, $\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\end{array}\right]$ For the coordinate position at every bar level line place, the first behavior horizontal ordinate, the ordinate that the second behavior is corresponding, dim1, dim2, dim3 ... for the quantity that every bar level line is put, s in formula (2) is added up respectively to elements all in vectorial S _hquantity, statistics is designated as m _h, and generate level line quantity vector M=[m _h]; Generate " level line quantity-altitude information " relation curve, wherein horizontal ordinate is vectorial S, and ordinate is level line quantity vector M; In conjunction with indium cylindricality looks feature, choosing specific between first peak and second peak in " level line quantity-altitude information " relation curve is highly the base altitude ZBasic of indium post Elevation Analysis;

Step 5: the level line coordinates matrix C solving base altitude ZBasic place _zbasic

C _zbasic＝contour(Z,[ZBasic ZBasic]) (3)

Level line coordinates matrix C _zbasicexpression is:

$\mathrm{Czbasic}=\left[\begin{array}{cccccccc}\mathrm{ZBasic}& \mathrm{xdata}\left(1\right)\mathrm{xdata}\left(2\right)& ...\mathrm{xdata}\left(\dim 1\right)& \mathrm{ZBasic}& \mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...& \mathrm{xdata}\left(\dim 2\right)& \mathrm{ZBasic}...\\ \dim 1& \mathrm{ydata}\left(1\right)\mathrm{ydata}\left(2\right)& ...\mathrm{ydata}\left(\dim 1\right)& \dim 2& \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...& \mathrm{ydata}\left(\dim 2\right)& \dim 3...\end{array}\right]---\left(4\right)$

With the ZBasic element value in formula for mark, isolate a series of level line coordinates matrix:

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 1\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 1\right)\end{array}\right]---\left(5\right)$

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 2\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 2\right)\end{array}\right]---\left(6\right)$

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 3\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 3\right)\end{array}\right]---\left(7\right)$

……；

Step 6: formula (5) is denoted as:

$\left[\begin{array}{c}\mathrm{Xdata}1\\ \mathrm{Ydata}1\end{array}\right]---\left(8\right)$

In formula, vectorial Xdata1, Ydata1 are:

Xdata1＝[xdata(1) xdata(2) ... xdata(dim1)] (9)

Ydata1＝[ydata(1) ydata(2) ... ydata(dim1)] (10)

Obtain vectorial Xdata1 respectively, in vectorial Ydata1 element maximal value and round and obtain Xdata1 _maxand Ydata1 _max, ask its minimum value respectively and round and obtain Xdata1 _minand Ydata1 _min, generate horizontal ordinate vector X _1s:

X _1S＝[Xdatal _minXdatal _min+1Xdatal _min+2…Xdatal _max] (11)

Ordinate vector Y _1s:

Y _1S＝[Ydatal _minYdatal _min+1Ydatal _min+2…Ydatal _max] (12)

Solve [X ₁, Y ₁]:

[X ₁,Y ₁]＝meshgrid(X _1s,Y _1s) (13)

Generate horizontal ordinate matrix X ₁, ordinate matrix Y ₁, in formula (13), meshgrid () is the function in Matlab7.0;

Step 7: obtain horizontal ordinate matrix X ₁, ordinate matrix Y ₁corresponding altitude information matrix Z ₁=[z1 _kl], Z ₁middle element meets z1 _kl=z _kl, wherein k, l are subscript, and have:

Xdatal _min<k<Xdatal _max，Ydatal _min<l<Ydatal _max

The method in step 4 is utilized to obtain matrix Z ₁corresponding " level line quantity-altitude information " relation curve, utilize this relation curve, in conjunction with the pattern feature on indium post surface, the approximate altitude value ZApro1 of the level line quantity chosen in curve near second peak to be the height of the assigned address of 1 be single indium post;

Step 8: the C solving single indium post approximate altitude value ZApro1 place _zapro1

C _zapro1＝contour(Z ₁,[ZApro1 ZApro1]) (14)

C in formula _zapro1expression is:

$\mathrm{Czapro}1=\left[\begin{array}{cc}\mathrm{ZApro}1& \mathrm{xdata}\left(1\right)\mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 1\right)\\ \dim 1& \mathrm{ydata}\left(1\right)\mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 1\right)\end{array}\right]---\left(15\right)$

With the ZApro1 value in formula for mark, have and only have an isocontour coordinates matrix

$\left[\begin{array}{cc}\mathrm{xdata}\left(1\right)& \mathrm{xdata}\left(2\right)...\mathrm{xdata}\left(\dim 1\right)\\ \mathrm{ydata}\left(1\right)& \mathrm{ydata}\left(2\right)...\mathrm{ydata}\left(\dim 1\right)\end{array}\right]---\left(16\right)$

Solve single indium post height height ₁:

height ₁＝mean(z1 _pq) (17)

Wherein p, q are subscript, meet xdata (1) <p<xdata (dim 1), ydata (1) <q<ydata (dim 1);

Step 9: to formula (6), formula (7) ... a series of level line coordinates matrix respectively repeats steps six, seven, eight, draws corresponding a series of indium post height value height ₁, height ₂, height ₃, generate indium post height value vector H,

H＝[height ₁,height ₂,height ₃…] (18)

Solve hist (H), obtain the statistic histogram of indium post height, wherein hist () is the function in Matlab7.0.