CN100372123C - CCD image sensor and high-precision linear dimension measuring device and measuring method thereof - Google Patents

Abstract

The invention relates to a CCD image sensor. The measuring device and measuring method of the sensor are used. The sensor is composed of a plurality of photodiode rows (1) arranged in a staggered manner, and each photodiode row (1) has corresponding charges arranged in a row. The transfer device (2), the photodiode row (1) is connected to the charge transfer device (2) through the read gate (3), the output of all the charge transfer devices (2) is connected to the same control circuit (4), and N photoelectric The diode row (1) is equidistantly staggered according to a distance H ₁ in one direction of the row, H ₁ =H/N, and H is the center distance between two adjacent photodiodes in the photodiode row (1). The sensor and the optical lens group form a measuring device. The measurement method is to place the measured object between the optical lens group and the sensor, measure the total length of the light-emitting diodes on the sensor that are not obviously sensitive due to the blocking of the measured object, and calculate the length of the measured object. .

Description

CCD image sensor and high-precision linear dimension measuring device and measuring method thereof

Technical field:

The present invention relates to a CCD image sensor, to a linear dimension measuring device using the sensor, and to a measuring method using the device.

Background technique:

In recent years, CCD image sensors have been widely used in non-contact measurement and computer vision imaging. In the measurement of CCD image sensor, improving the accuracy of its measurement is a problem that must be considered. The solution can be realized from two aspects: on the one hand, the resolution of imaging can be directly improved from the CCD hardware structure; measurement accuracy. At present, effective methods for improving measurement accuracy are proposed from different angles, but none of them can break through the influence of CCD pixel spacing, so it is difficult to make a qualitative leap in measurement accuracy.

First, analyze the structure of CCD image sensor.

FIG. 1 is a schematic plan view of a conventional CCD image sensor 100A.

It mainly consists of a photodiode row 102 composed of a plurality of photodiodes arranged in a row and a charge transfer device 101 corresponding to the photodiode row and arranged in a row. Each photodiode in photodiode row 102 outputs a signal charge to charge transfer device 101 through read gate 103 . The charge on the charge transfer device 101 is then transferred to the relevant signal processing circuit, finally forming the output signal of the CCD image sensor.

FIG. 2 is a schematic plan view of another conventional CCD image sensor 200B of a dual CCD structure proposed in Japanese Patent Application Laid-Open No. 11-164087.

The image sensor includes two rows of charge transfer devices 201a and 201b, and their corresponding row 202 of photodiodes. The rows of photodiodes are divided into two groups, the group 202a ejecting charge to the charge transfer device 201a and the group 202b ejecting charge to the charge transfer device 201b. In such an arrangement, the charge transfer devices 201a and 201 may be arranged equidistantly from the charge transfer device 201 of the single CCD structure shown in FIG. 1 . This enables a dual-CCD image sensor to have twice as many photodiodes as a single-CCD image sensor while reducing the size of the photodiodes by half, without the need to fabricate smaller-sized charge-transfer devices.

3 is a schematic plan view of another CCD image sensor 300C proposed in Japanese Patent Application Laid-Open No. 2001-203342 including two photodiode rows arranged in a staggered arrangement.

The CCD image sensor 300C shown in FIG. 3 includes two separate photodiode rows 302a and 302b, and their corresponding first charge transfer device 301a and second charge transfer device 301b. The two photodiode rows 302a and 302b are arranged alternately with half the pitch between two adjacent photodiodes. Similar to the dual-CCD image sensor 200B shown in FIG. 2, this structure allows the CCD image sensor 300C to have photodiodes twice in number than the single-CCD type CCD image sensor without the need to manufacture a smaller-sized charge transfer device. . In addition, the CCD image sensor 300C has an obvious advantage over the dual-CCD image sensor 200B: larger photodiodes can be manufactured, which ensures a higher signal-to-noise ratio and a wider dynamic range of the sensor.

FIG. 4 is a schematic plan view of a CCD image sensor 400D proposed in Chinese Patent Application Publication No. 200310119527.8 including a plurality of rows of photodiodes arranged in a staggered arrangement.

The CCD image sensor 400D shown in FIG. 4 includes a first diode row 402 and a second diode row 402 . And each row of diodes is divided into two parts according to the order of odd and even, respectively 402a, 402b and 402c, 402d; the first charge transfer device 401a, the second charge transfer device 401b, the third charge transfer device 401c and the fourth charge transfer device 401d . The first diode row and the second diode row are alternately arranged at an interval of half of the photodiodes in the first diode row along the extending direction of the photodiode arrangement. The first charge transfer device transfers and receives the signal charge of the Kth photodiode in the first diode row, K is an odd number; the second charge transfer device transfers and receives the signal charge of the Lth photodiode in the first diode row charge, L is an even number; the third charge transfer device transfers and receives the signal charge of the Kth photodiode in the second diode row, K is an odd number; the fourth charge transfer device transfers and receives the signal charge of the Kth photodiode in the second diode row The signal charge of L photodiodes, L being an even number; the CCD image sensor 400D can arrange the photodiodes at a higher density without making a smaller-sized charge transfer device, and can improve Distortion in charge transfer due to track length.

Of the above four types of CCD image sensors, the first one is the basic type, and the other three are modifications to the first structure, the purpose of which is to arrange more photodiodes in a limited area.

FIG. 5 is a schematic diagram of a photodiode arrangement plane of a super CCD image sensor 500E launched by Fuji Corporation of Japan. In the article "Super CCD Principle" of "Sensor Technology" No. 4, 2003, its structural principle was analyzed in detail.

In the super CCD image sensor 500E, octagonal photodiodes are used to replace ordinary rectangular diodes, and the photodiodes are arranged at an angle of 45° to form a honeycomb arrangement structure. Super CCD was developed in 1999. The octagonal photodiode and honeycomb pixel arrangement greatly improved the space efficiency of the photodiode in each pixel unit. This brings numerous additional benefits, such as higher sensitivity, higher signal-to-noise ratio, and wider dynamic range than conventional CCDs with the same number of pixels. Its biggest feature is that the photodiodes are arranged at an angle of 45°. This arrangement and the RGB three-color mode can just make full use of the signals obtained by adjacent photodiodes. In data processing, the super CCD image sensor uses three R, G, and B photosensitive diodes to form a color pixel. Although the photosensitive unit has only three primary colors, each unit is multiplexed 6 times. Therefore, the super CCD image sensor photosensitive Although the diode data has not changed, the number of pixels generated after processing is twice that of ordinary CCD image sensors. Super CCD image sensor has obvious advantages in digital color imaging and is widely used, but it has no obvious advantages in high-precision measurement applications.

The following analyzes the method of improving the measurement accuracy of the CCD image sensor measurement system by changing other devices other than the CCD image sensor in the measurement system and using data processing algorithms.

FIG. 6 is a schematic diagram of a method for improving measurement accuracy by projecting and enlarging the measured object by using the magnifying optical measurement system.

The enlarged optical measurement system shown in FIG. 6 mainly includes a light source 5 , an optical lens group 6 , a measured object 7 , and a CCD image sensor 8 . In this method, the measured object is enlarged under the irradiation of divergent light, and the enlarged shadow is photosensitive and imaged on the CCD image sensor, so that the CCD image sensor measures the enlarged projection of the measured object. Through such processing, the error (unilateral) of the projection measurement of the measured object is still determined by the pixel pitch of the CCD image sensor (the distance between adjacent photodiodes), and the maximum will not exceed one pixel pitch. By dividing the measured data by the optical magnification, the measured value of the measured object can be obtained, which can basically reduce the measurement error by the corresponding optical magnification. For example, for an object with a length of 1.0145mm, use a linear array CCD image sensor with a pixel pitch of 10μm to measure, if the optical magnification is 10, then the projected length of the measured object is 10.145mm. Assuming no other edge processing methods are used, the projected length of the measured object may be 10.14mm or 10.15mm, divided by the optical magnification of 10, the measured value of the measured object is 1.014mm or 1.015mm, and the maximum error is 0.0005mm. If the 1:1 optical imaging system is directly used for measurement, the maximum measured value may be 1.02mm, the minimum may be 1.01mm, and the maximum error is 0.0055mm. It can be seen that the optical magnification imaging system can basically reduce the measurement error by the corresponding optical magnification.

In "Journal of Sichuan University" No. 5, 2001, "A new method to improve the measurement accuracy of CCD" introduced a fuzzy imaging method. The method is a method for improving the measurement accuracy of the CCD by increasing the components of the measurement system.

The feature of the fuzzy imaging method is to simply improve the signal acquisition part of the CCD measurement system. Specifically, a suitable aperture diaphragm is added to the lens group to blur the clear edge of the object after being imaged by the CCD, and then use the diaphragm to Fit the measurement results with known conditions such as the size of the lens and the focal length of the lens to obtain accurate edge information of the measured object. This method is difficult to accurately determine the degree of improvement in measurement accuracy theoretically, but from the experimental data, it has a significant effect within a certain range, such as using a linear array CCD with a pixel pitch of 14 μm to measure a standard sample The diameter is measured, and the measurement error can basically be controlled within 5 μm.

In the article "An Attempt to Improve CCD Resolution" in "Modern Metrology Technology" No. 3, 1997, an edge fitting method for CCD imaging was introduced.

The edge fitting method is based on the fact that in the actual measurement system, due to the diffraction effect of light, as well as the existence of spherical aberration, aberration and focusing error of the imaging system, as well as the influence of noise, the output of the CCD image sensor is a slope-like curve mixed with noise. The edge fitting method is to obtain a more accurate edge position by performing algorithmic processing on the signal curve actually output by the CCD image sensor. Its basic design idea is: first smooth (fit) the original grayscale image, and then perform gradient processing on the smoothed image, and then find the position of the point with the largest gradient value on the gradient image, then use the The position of the point is defined as the position of the edge point. Similarly, it is difficult to accurately determine the improvement value of its measurement accuracy theoretically, but from the statistical data point of view, reasonable use of it can at least double the measurement error.

Whether it is to use an enlarged optical measurement system or an edge data processing method, the data measurement accuracy of the CCD image sensor can be improved. However, all these processing methods are currently adopted under the condition that they cannot break through the influence of the pixel pitch of the CCD image sensor, so it is difficult to make a qualitative leap in measurement accuracy.

What was invented:

The object of the present invention is to provide a CCD image sensor which is particularly suitable for measuring linear dimensions requiring high precision along a certain direction.

Another object of the present invention is to provide a high-precision linear dimension measuring device including the above-mentioned CCD image sensor.

Another object of the present invention is to provide a high-precision measurement method using the above-mentioned device.

The present invention is achieved like this:

The CCD image sensor of the present invention is composed of a plurality of photodiode rows 1 arranged in a staggered manner, each photodiode row 1 has a corresponding charge transfer device 2 arranged in a row, and the photodiode row 1 connects with the charge transfer device through a read gate 3 2 connections, the output of all charge transfer devices 2 is connected to the control circuit 4, the output of the control circuit 4 is connected to the signal processing circuit, N photodiode rows 1 are arranged equidistantly at a distance H ₁ in one direction of the row, H ₁ =H/N, H is the distance between the centers of two adjacent electric diodes in photodiode row 1, N is the number of photodiode rows, N>2.

Signal processing circuit is made up of image acquisition device and microcomputer, the input of the amplification filter circuit of image acquisition device connects the output of control circuit 4, the output of amplification filter circuit connects A/D conversion circuit, A/D conversion circuit is respectively connected with the memory of microcomputer and the output of control circuit 4. The microprocessor is connected, and the microprocessor is connected with the memory and the output terminal.

The light source 5, optical lens group 6, measured object 7 and CCD image sensor 8 are sequentially arranged. The edge of the measured object 7 is projected on one or two CCD image sensors 8 on the same plane, and the edge projection line is consistent with the CCD image sensor 8. The photodiode row axes of the image sensor 8 are perpendicular or crossed.

Method of the present invention, its step is as follows:

(1) The measured object 7 is placed between the optical lens group 6 and the CCD image sensor 8, the projected line of the measured object is vertically intersected with the photodiode row axis of the CCD image sensor 8, and its two edge projections are all on the sensor 8 ,

(2) Determine that the edge of the measured object falls between the nth and n+1 pixels on the first photodiode row of the CCD image sensor 8, the nth pixel has no obvious light sensitivity, and the n+1th pixel has obvious light sensitivity , and measure the number m ₁ of all the photodiode rows that meet the above conditions. According to the same method, it is determined that all n pixels that fall on the CCD image sensor 8 on the other side of the object under test have no obvious light sensitivity, and n+1 pixels have obvious light sensitivity. photosensitive line number m ₂ ,

(3) Measure the total length a between the photodiodes on the CCD image sensor 8 that are not obviously sensitive to light due to the occlusion of the measured object 7, a=H×(n ₂ -n ₁ ), n ₁ is the first light sensitive The number of array element columns corresponding to the photodiode, n ₂ is the number of array element columns corresponding to the last photodiode that is not obviously sensitive to light, and H is the pixel spacing of the sensor along the row direction.

(4) Calculate the length L of the measured object:

L=[a+(H/N)(m ₁ -1)+(H/N)(m ₂ -2)]/M

H is the pixel pitch of the CCD image sensor along the row direction

N is the total number of rows of photodiodes

M is the optical magnification of the measurement system.

Method of the present invention, its step is as follows:

(1) The measured object 7 is placed between the optical lens group 6 and the CCD image sensor 8, the projection line of the measured object is perpendicular to the photodiode line axis of the CCD image sensor 8, and one edge is projected on the sensor 8,

(2) Determine that the edge of the measured object 7 is projected between the nth pixel and the n+1th pixel on the first photodiode row of the sensor 8, the nth pixel has no obvious light sensitivity, and the n+1th pixel is obvious Sensitize, and measure the number of photodiode rows m that meet the above conditions,

(3) measure the total length a between the photodiodes without obvious light sensitivity due to the blocking of the measured object 7 on the measurement CCD image sensor 8, measure the distance b between the end of the measured object away from the sensor 8 and the sensor 8 near the end of the measured object, b is calibrated with a micrometer screw calibrator with a resolution smaller than 1 μm, a=H×(n ₂ -n ₁ ), H is the pixel pitch of the sensor along the row direction, and n ₁ is the photodiode corresponding to the first photosensitive photodiode The number of array element columns, n ₂ is the number of array element columns corresponding to the last photodiode that is not obviously sensitive to light,

(4) Calculate the length L of the measured object:

L=[a+b+(H/N)(m-1)]/M or

L=[a+b+(H/N)(m-2)]/M

N is the total number of rows of photodiodes,

M is the optical magnification.

Method of the present invention, its step is as follows:

(1) The measured object is placed between the optical lens group 6 and the CCD image sensor 8, the projected line of the measured object is vertically intersected with the photodiode row axes of the two sensors 8 on the same plane, and its two edges are projected on the on different sensors 8,

(2) Determine that the side edge of the measured object falls between the nth and n+1th pixels on the first photodiode row of the first sensor 8, the n pixel has no obvious light sensitivity, and the n+1 pixel has obvious light sensitivity, measure Find out the photodiode line number m ₁ that the first sensor has the above-mentioned conditions, and according to the same method, measure the number of photodiodes projected on the second sensor (8) on the other side edge of the measured object so that the second sensor 8 has the above-mentioned conditions. number m ₂ ,

(3) Measure the distance b between the two sensors 8 close to each other. The lengths b ₁ , b ₂ , and b of the photodiodes on the first and second sensors 8 that have no obvious light sensitivity due to the occlusion of the object to be measured are calibrated with a micrometer screw calibrator, b ₁ , b ₂ are equal to the value of H×(n ₂ -n ₁ ), H is the pixel spacing along the row direction of the sensor, n ₁ is the number of array element columns corresponding to the first photodiode that is not obviously sensitive to light, and n ₂ is The number of array element columns corresponding to the last photodiode that is not obviously sensitive to light,

(4) Calculate the length L of the measured object

L=[b+b ₁ +b ₂ +(H/N)(m ₁ -1)+(H/N)(m ₂ -2)]/M

N = total number of photodiode rows for the two sensors (8).

M is the optical magnification.

Description of drawings:

FIG. 1 is a schematic plan view of a conventional CCD image sensor.

Fig. 2 is a schematic plan view of a conventional dual-CCD type CCD image sensor.

3 is a schematic plan view of a CCD image sensor including two photodiode rows arranged in a staggered arrangement.

4 is a schematic plan view of a CCD image sensor including a plurality of photodiode rows arranged in a staggered arrangement.

FIG. 5 is a schematic plan view of the photodiode arrangement of the super CCD image sensor.

Fig. 6 is a schematic diagram of an enlarged optical measurement device.

FIG. 7 is a schematic plan view of a CCD image sensor in which three columns of diodes are arranged equidistantly and staggered.

FIG. 8 is a schematic plan view of a CCD image sensor in which seven rows of diodes are arranged equidistantly and staggered.

FIG. 9 is a flow chart of how to judge the edge profile of the measured object during the measurement process of the CCD image sensors arranged equidistantly and staggered.

Fig. 10 is a schematic diagram of a CCD image sensor optical imaging measuring device according to the first embodiment of the present invention.

Fig. 11 is a schematic diagram of a CCD image sensor optical imaging measurement device according to the second embodiment of the present invention.

Fig. 12 is a schematic diagram of a CCD image sensor optical imaging measurement device according to a third embodiment of the present invention.

FIG. 13 is a schematic diagram of a structure in which a plurality of traditional linear CCD image sensors are arranged equidistantly and staggered along the length direction thereof.

Figure 14 is a schematic diagram of the control circuit.

Detailed ways:

Example 1:

As shown in FIG. 10 , it is a schematic diagram of a CCD image sensor optical imaging measurement device according to the first embodiment of the present invention. The first embodiment of the present invention mainly includes a photodiode equidistant staggered CCD image sensor 8 , an optical lens group 6 , a light source 5 , a signal processing circuit 4 and a measured object 7 . Each photodiode row of the CCD image sensor is equidistantly staggered, and the number of photodiode rows is determined by the measurement accuracy and the distance between two adjacent photodiodes along the row direction. The magnification of the optical imaging system can be 1, or any multiple greater than 1. The feature of the first embodiment of the present invention is: the rows of photodiodes in the CCD image sensor are arranged equidistantly and staggered, and the measured object is measured bilaterally, that is, both ends of the measured object are completely imaged on the CCD image sensor. The center of the measured object can be placed on the central section of the CCD image sensor, or placed in other appropriate places, as long as the measured object can be completely imaged on the CCD image sensor. The measurement data can be directly obtained by counting the number of blocked photodiodes on the CCD image sensor.

Figure 14 is a schematic diagram of the processing process of the CCD image signal in the measuring device.

Due to the occlusion of the light by the measured object, different photosensitive signals are generated on the photosensitive diodes of the CCD image sensor. The signal is used as the output of the CCD, and after amplification, filtering and A/D conversion, it can be directly sent to the CPU for calculation. It can also be sent to memory and processed later. The processed results can be output through monitors, printers, etc.

CCD output signal amplification filtering and A/D conversion circuit, currently there are relatively mature data acquisition cards that can complete the processing work in this area, such as the CA-MPE-1000 image acquisition card. The CPU, memory and output can be completely completed by a PC. Therefore, the processing of the CCD output signal can be carried out by using the existing data acquisition card and PC. It is only necessary to compile the processing program according to the corresponding process.

In the process of data processing, the electrical signals output by the photodiodes on the first row of the CCD image sensor are stored in the first row of the memory array in turn, and in this way, the photodiodes on the other rows of the CCD image sensor are respectively The photosensitive signals are stored in the corresponding positions of the array in turn. Assume that the array is D(i, j), which is a two-dimensional array, and the number of rows and columns of the array are the same as those of the CCD image sensor. And the elements therein correspond one-to-one to the photosensitive diodes on the CCD image sensor. Therefore, during data processing, each element in the array corresponds to each photodiode on the CCD image sensor.

The measurement result analysis process can be briefly described as follows: first read the first row of elements in the array D, read from the first element of the first row until the last element of the first row, and find that the nth element is obviously sensitive to light (not Obvious photosensitive), and the n+1th element is not obvious photosensitive (obvious photosensitive) of the two photosensitive diode positions, and calibrate it. Then the edge to be measured falls between the pixels corresponding to the nth and n+1th elements of the array on the photodiode row. Then, read the second row in the array D, find the array elements with the same number of photodiode columns that have been calibrated in the first row, and judge their photosensitive conditions. If they are the same as the first row, continue to read the next row of data. Otherwise, calibrate m ₁ (m ₂ ), if the current row number is i, then m ₁ (m ₂ )=i-1. Finally, the measured edge position is calculated through data processing.

To specifically judge whether a certain pixel is obviously sensitive to light, the simplest binary method can be used. That is, a threshold is set, and when the photosensitive signal voltage of the pixel is higher than the threshold, the pixel is considered to be obviously photosensitive; otherwise, the pixel is considered not to be obviously photosensitive. For example, if the electrical signal output by the CCD is processed and converted into a standard voltage of 0-12V, the threshold value can be set to 5V. When the photosensitive signal of a pixel is higher than 5V, the pixel is obviously sensitive to light, otherwise , it is considered that the pixel is not obviously sensitive to light. Specifically, in the process of use, the threshold value can be determined through multiple tests and repeated adjustments to obtain a more reasonable value.

For the determination of a in Example 1:

In the first row of elements of the array D, if the number of array element columns corresponding to the first non-obvious photosensitive photodiode is n ₁ , and the number of array element columns corresponding to the last non-obvious photosensitive photodiode is n ₂ , H is the distance between the centers of two adjacent photodiodes in the photodiode row, then:

a=H×(n ₂ -n ₁ )

L=[a+(H/N)(m ₁ -1)+(H/N)(m ₂ -2)]/M

H is the pixel pitch of the CCD image sensor along the row direction.

N is the total number of rows of photodiodes, and M is the optical magnification.

Example 2:

As shown in FIG. 11 , it is a schematic diagram of a CCD image sensor optical imaging measuring device according to the second embodiment of the present invention. The second embodiment of the present invention mainly includes a photodiode equidistant staggered CCD image sensor 8 , an optical lens group 6 , a light source 5 , a signal processing circuit 4 and a measured object 7 . Where b represents the distance between the end of the measured object away from the CCD image sensor and the end of the CCD image sensor close to the measured object during the measurement process. In the measurement device, the end of the measured object far away from the CCD image sensor is placed on a reference plane, so b is constant constant. In the second embodiment of the present invention, the rows of photodiodes of the CCD image sensor are equidistantly staggered, and the number of rows of photodiodes is determined by the measurement accuracy and the distance between two adjacent photodiodes along the row direction. The magnification of the optical imaging system can be 1, or any multiple greater than 1. The characteristics of the second embodiment of the present invention are: the rows of photodiodes in the CCD image sensor are equidistantly staggered, and the measured object is unilaterally measured, that is, only part (one end) of the measured object is imaged on the CCD image sensor, which can be compared High-precision measurement of large-sized measured objects. The measurement data is obtained from the b value in Figure 11 and the number of blocked photodiodes on the CCD image sensor, and the projected length of the measured object on the CCD image sensor.

For the determination of m, a and b in the unilateral measurement of embodiment 2:

In Example 2, b is first calibrated and calibrated using a micrometer screw calibrator with a resolution of less than 1 μm.

First read the first row of elements in the array D, first read from the first element of the first row until the last element of the first row, and find that the nth element is not obviously photosensitive, while the n+1th element is obviously photosensitive The position of those two photodiodes and calibrate it. Then, read the second row in group D, find the array elements with the same number of calibrated photodiode columns in the first row, and judge their photosensitive conditions. If they are the same as the first row, continue to read the next row of data. Otherwise, calibrate m, if the current line number is i, then:

m=i-1

In the element in the first row of the array D, if the number of array element columns corresponding to the first photodiode that is not obviously sensitive to light is n ₁ , and the number of column elements in the array corresponding to the last photodiode that is not obviously sensitive to light is n ₂ , H is the distance between the centers of two adjacent photodiodes in the photodiode row, then:

a=H×(n ₂ -n ₁ )

L=[a+b+(H/N)(m-1)]/M

Example 3:

As shown in FIG. 12 , it is a schematic diagram of a CCD image sensor optical imaging measurement device according to the third embodiment of the present invention. The third embodiment of the present invention mainly includes two CCD image sensors 8a, 8b equidistantly staggered with photodiodes, an optical lens group 6, a light source 5, a signal processing circuit 4 and a measured object 7. Where b represents the distance between the two ends of the CCD image sensor 8 a and the CCD image sensor 8 b that are close to each other during the measurement process, which is a certain value. In the measuring device, the CCD image sensor 8a and the CCD image sensor 8b are arranged along the same straight line. In the third embodiment of the present invention, the rows of photodiodes of the CCD image sensor are equidistantly staggered, and the number of rows of photodiodes is determined by the measurement accuracy and the distance between two adjacent photodiodes along the row. The magnification of the optical imaging system can be 1, or any multiple greater than 1. The third embodiment of the present invention is characterized in that: the rows of photodiodes in the CCD image sensor are arranged equidistantly and staggered, and bilateral measurement is performed on the measured object, and high-precision measurement can be performed on larger-sized measured objects. The measurement data is obtained from the b value in Figure 12 and the number of blocked photodiodes on the CCD image sensor, and the projected length of the measured object on the CCD image sensor is specifically judged.

For the determination of m ₁ (m ₂ ), b, b ₁ , b ₂ in the unilateral measurement of embodiment 3:

In Example 3, b is first calibrated and calibrated using a micrometer screw calibrator with a resolution of less than 1 μm.

The light-sensing data of the photodiodes of the CCD image sensor 8a are stored in the array D ₁ , and the light-sensing data of the photodiodes of the CCD image sensor 8b are stored in the array D ₂ .

First read the first row of elements in the array D ₁ , first read from the first element of the first row until the last element of the first row, and find that the nth element is not obviously photosensitive, while the n+1th element is obvious The position of the two photosensitive diodes that are sensitive to light, and calibrate it. Then, read the second row in group D ₁ , find the array elements with the same number of calibrated photodiode columns in the first row, and judge their photosensitive conditions. If they are the same as the first row, continue to read the next row of data , otherwise calibrate m ₁ , if the current row number is i, then:

m ₁ =i-1

Then read the first row of elements in the array D ₂ , first read from the first element of the first row until the last element of the first row, and find that the nth element is not obviously photosensitive, while the n+1th element is obvious The position of the two photosensitive diodes that are sensitive to light, and calibrate it. Then, read the second row in the array D ₂ , find the array elements with the same number of photodiode columns that have been calibrated in the first row, and judge their photosensitive conditions. If they are the same as the first row, continue to read the next row of data , otherwise calibrate m ₂ , if the current row number is j, then:

m ₂ =j-1

In the first row of elements of the array D ₁ , if the total number of array elements corresponding to the photodiodes that are not obviously sensitive to light is n ₁ , then:

b ₁ =H×n ₁ .

In the first row of elements of the array D ₂ , if the total number of array elements corresponding to the photodiodes that are not obviously sensitive to light is n ₂ , then:

b ₂ =H×n ₂ .

L=[b+b ₁ +b ₂ +(H/N)(m ₁ -1)+(H/N)(m ₂ -2)]/M

Example 4:

As shown in FIG. 13 , it is a schematic diagram of a structure in which a plurality of traditional linear CCD image sensors are arranged equidistantly and staggered along its length direction. The staggered distance between the linear CCD image sensors along the row direction is determined by the measurement accuracy and the distance between two adjacent photodiodes along the row direction. Such a plurality of linear CCD image sensors that are equidistantly staggered is called a linear CCD image sensor. Image sensor group. In Fig. 13, line array CCD image sensor group 8 is made up of line array CCD image sensor 8a, 8b, 8c, 8d, the number of line array CCD image sensor that forms line array CCD image sensor group 8 can be any greater than 2 as required integer. The CCD image sensor parts in the first embodiment, the second embodiment, and the third embodiment of the present invention are replaced by the linear array CCD image sensor group respectively, so as to form a line similar to the first embodiment and the second embodiment of the present invention along the row direction. 1. The structure in which photodiode columns are equidistantly staggered in the third embodiment. The magnification of the optical imaging system can be 1, or any multiple greater than 1. The fourth embodiment of the present invention is characterized in that adjacent CCD image sensors in the line array CCD image sensor group are equidistantly staggered. The CCD image sensors in the third embodiment of the present invention can be replaced by the equidistant staggered linear array CCD image sensor groups in the fourth embodiment of the present invention to obtain a corresponding measuring device.

Claims (6)

1. ccd image sensor, constitute by staggered a plurality of photodiodes capable (1), each photodiode capable (1) has correspondence to be arranged into the charge transfer device (2) of delegation, photodiode capable (1) is connected with charge transfer device (2) by reading grid (3), the output of all charge transfer devices (2) is connected with control circuit (4), the output of control circuit (4) is connected with signal processing circuit, it is characterized in that on the direction that N photodiode capable (1) be expert at by distance H ₁Equidistantly staggered, H ₁=H/N, H are the distance at adjacent two electric diode centers of photodiode capable (1), and N is the photodiode line number, N＞2.

2. ccd image sensor according to claim 1, it is characterized in that said signal processing circuit is made of image collecting device and microcomputer, the output of the input connection control circuit (4) of the amplification filtering circuit of image collecting device, the output of amplification filtering circuit connects the A/D change-over circuit, the A/D change-over circuit is connected with microprocessor with the memory of microcomputer respectively, and microprocessor is connected with memory, outlet terminal.

3. high accuracy linear dimension measuring device that comprises claim 1 or 2 described ccd image sensors, it is characterized in that constituting by layout successively by light source (5), optical lens group (6), measurand (7) and ccd image sensor (8), the edge projection of measurand (7) is positioned on the conplane ccd image sensor (8) at one or two, the capable axis normal of photodiode of edge projection line and ccd image sensor (8) or intersect.

4. method based on the measurement measurand linear dimension of the described device of claim 3, its step is as follows:

(1) measurand (7) is placed between optical lens group (6) and the ccd image sensor (8), the capable intersect vertical axis of photodiode of measurand projection line and ccd image sensor (8), and its two edges projection is all on transducer (8),

(2) determine that the measured object edge drops between first photodiode of ccd image sensor (8) on capable n and n+1 the pixel, n pixel do not have obvious sensitization, n+1 the obvious sensitization of pixel, and measure all photodiode line number m that possesses above-mentioned condition ₁, determine that according to same method all n pixel that measured object opposite side edge drops on ccd image sensor (8) do not have obvious sensitization, the line number m of n+1 the obvious sensitization of pixel ₂,

(3) measure ccd image sensor (8) and go up the total length a between the photodiode that does not have obvious sensitization of blocking because of measured object (7), a=H * (n ₂-n ₁), n ₁Be the array element columns of the photodiode correspondence of first not obvious sensitization, n ₂Be the pairing array element columns of the photodiode of last not obvious sensitization, H is the pixel spacing that transducer follows direction.

(4) calculate the measurand length L:

L＝[a+(H/N)(m ₁-1)+(H/N)(m ₂-2)]/M

H is the pixel spacing that ccd image sensor follows direction

N is total line number of photodiode

M is the measuring system optical magnification.

5. method based on the linear dimension of the measurement measured object object of the described device of claim 3, its step is as follows:

(1) measurand (7) is placed between optical lens group (6) and the ccd image sensor (8), the capable axis normal of photodiode of the projection line of measurand and ccd image sensor (8), and the one edge projection is on transducer (8),

(2) determine between n pixel and n+1 pixel of measurand (7) edge projection on first photodiode of transducer (8) is capable, n pixel do not have obvious sensitization, n+1 the obvious sensitization of pixel, and measure the photodiode line number m that possesses above-mentioned condition

(3) measure ccd image sensor (8) and go up the total length a between the photodiode that does not have obvious sensitization of blocking because of measured object (7), measure the distance b of measurand away from an end and close measurand one end of transducer (8) of transducer (8), b demarcates a=H * (n with the little 1 μ m microspindle calibrator of resolution ₂-n ₁), H is the pixel spacing that transducer follows direction, n ₁Be the array element columns of the photodiode correspondence of first not obvious sensitization, n ₂Be the array element columns of the photodiode correspondence of last not obvious sensitization,

(4) calculate the measurand length L:

L=[a+b+ (H/N) is (m-1)]/M or

L＝[a+b+(H/N)(m-2)]/M

N is total line number of photodiode,

M is an optical magnification.

6. method based on the measurement measurand linear dimension of the described device of claim 3, its step is as follows:

(1) measurand is placed between optical lens group (6) and the ccd image sensor (8), the capable intersect vertical axis of photodiode of measurand projection line and two transducers (8) in the same plane, and its two edges are projected on the different transducer (8)

(2) determine that measurand one lateral edges drops between first photodiode of first sensor (8) n and n+1 pixel on capable, the n pixel does not have obvious sensitization, the photodiode line number m that first sensor possesses above-mentioned condition is measured in the obvious sensitization of n+1 pixel ₁,, measure the opposite side edge of measurand and count m in the photodiode institute that the last projection of second transducer (8) makes second transducer (8) possess above-mentioned condition according to Same Way ₂,

(3) measure two sensors (8) mutually near the distance b of end, first, second transducer (8) is gone up the photodiode length b that does not have obvious sensitization because of blocking of measured object ₁, b ₂, b demarcates with the microspindle calibrator, b ₁, b ₂Equal H * (n ₂-n ₁) value, H is that transducer follows direction pixel spacing, n ₁Be the array element columns of the photodiode correspondence of first not obvious sensitization, n ₂Be the array element columns of the photodiode correspondence of last not obvious sensitization,

(4) calculate the measurand length L

L＝[b+b ₁+b ₂+(H/N)(m ₁-1)+(H/N)(m ₂-2)]/M

N=is total photodiode line number of two transducers (8).

M is an optical magnification.

Images