WO2006108137A2

WO2006108137A2 - Glass inspection systems and methods for using same

Info

Publication number: WO2006108137A2
Application number: PCT/US2006/013012
Authority: WO
Inventors: Leon R. Zoeller, Iii
Original assignee: Corning Incorporated
Priority date: 2005-04-06
Filing date: 2006-04-06
Publication date: 2006-10-12
Also published as: TW200706859A; TWI360652B; KR20070121820A; WO2006108137A3; CN101175986B; CN101175986A; EP1866625A4; EP1866625A2; JP2008536127A

Abstract

Several different inspection systems (100) and methods are described herein that identify defects (e.g. inclusions, occlusions, scratches, stains, blisters, cords or other imperfections associated with surface discontinuities or materal non-homogeneities) on or within a lass sheet 110 .

Description

GLASS INSPECTION SYSTEMS AND METHODS FOR USING SAME

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Application Serial No. 60/669,171 filed on April 6, 2005 and entitled "Glass Inspection System and Methods for Using Same" which is incorporated by reference herein in.

BACKGROUND OF THE INVENTION

Field of the Invention

The present invention relates in general to inspection systems and methods that are used to identify defects on or within a glass sheet.

Description of Related Art

Manufacturers of glass sheets are always trying to design a new and improved inspection system that can be used to identify defects (e.g., inclusions, onclusions, scratches, stains, blisters, cords or other imperfections associated with surface discontinuities or material non- homogeneities) on or within a glass sheet (e.g., liquid crystal display (LCD) glass substrate) . Several new and improved inspection systems are the subject of the present invention. BRIEF DESCRIPTION OF THE INVENTION

The present invention includes several different embodiments of inspection systems and methods for identifying defects (e.g., inclusions, onclusions, scratches, stains, blisters, cords) on or within a glass sheet. In one embodiment, the inspection system includes an illuminator, a lens and a line-scan sensor. The illuminator emits a light beam and the lens receives the light beam and then emits a parallel light beam through a portion of a glass sheet. The line-scan sensor then receives the parallel light beam that passed through the glass sheet and is able to focus on the imperfections in the glass sheet without needing to have another lens placed between the line- scan sensor and the glass sheet.

BRIEF DESCRIPTION OF THE DRAWINGS

A more complete understanding of the present invention may be had by reference to the following detailed description when taken in conjunction with the accompanying drawings wherein:

FIGURES IA- IF show six diagrams associated with a first embodiment of the inspection system in accordance with the present invention;

FIGURES 2A-2C show three diagrams associated with a second embodiment of the inspection system in accordance with the present invention;

FIGURES 3A-3E show five diagrams associated with a third embodiment of the inspection system in accordance with the present invention; FIGURES 4A-4C show three diagrams associated with a fourth embodiment of the inspection system in accordance with the present invention; FIGURES 5A-5C show three diagrams associated with a fifth embodiment of the inspection system in accordance with the present invention;

FIGURES 6A- 6D show four diagrams associated with a sixth embodiment of the inspection system in accordance with the present invention; and

FIGURES 7A- 7D show four diagrams associated with a seventh embodiment of the inspection system in accordance with the present invention.

DETAILED DESCRIPTION OF THE DRAWINGS

Referring to FIGURES IA-IF, there are shown six diagrams associated with a first embodiment of the inspection system 100 in accordance with the present invention. FIGURE IA shows the inspection system 100 which includes a diode laser 102 that generates a laser line 104 which passes ^' through a cylindrical lens 106 that refracts all of the rays in the laser line 104 into a beam of parallel light 108 which is transmitted through a glass sheet 110 and received by a line-scan sensor 112. An important aspect of the inspection system 100 is that the line- scan sensor 112 is able to focus on the imperfections in the glass sheet 110 without needing to have another lens placed between the line- scan sensor 112 and the glass sheet 110. In this example, the distance between the cylindrical lens 106 and the sensor 112 is approximately

4". The width of the beam of parallel light beam 108 is

3-5". And, the glass sheet 110 can vary + 1 inch in position between the cylindrical lens 106 and the sensor 112.

The inspection system 100 is a marked improvement over the traditional inspection system which required a lens to be precisely placed between the glass sheet 110 and the sensor 112 in order to focus on a defect in the glass sheet 110. For instance, a traditional inspection system that could detect imperfections on the order of 1 to 200 microns needed to have a depth of field less than a couple of millimeters. In contrast, the inspection system 100 of the present invention has an equivalent depth of field in the range of inches. This is because the inspection system 100 relies on light rays 104 traveling directly from the laser 102 through the collimating lens 106 (the only purpose of which is to make the light rays go parallel to each other) to a small sensor element in the sensor 112. And, if any small object such as a defect disturbs the light path then this disturbance is captured by the sensor 112. The disturbance can occur at any point in the light path between the laser 102 and the sensor 112. Thus, the inspection system 100 can detect and measure an imperfection with a much more relaxed sensor to object distance when compared to the traditional inspection system. FIGURES IB- IE show different imperfection images associated with different glass sheets 110 that were scanned by the inspection system 100. To obtain each image, the sensor 112 would output a signal which was quantized and input into a computer (not shown) where it was analyzed with image processing algorithms so it could be displayed in graphical form. Looking at the images generated it should noted that the height of the signal is caused by the imperfection. The imperfection is a small inclusion and in this series of experiments it was a small particle of platinum, zirconium, stainless steel or some other contaminant. To generate these images, the laser 102 had to produce a light line 104 which was highly coherent so as to enable a fresnel effect . The fresnel effect is where the light energy goes around a dark imperfection and forms into a peak which can be higher than the light 104 produced by the laser 102. As such, the fresnel effect produces a very high signal to noise ratio which allows small imperfections to be easily detected.

FIGURE IF is a graph that shows the variation in size of a defect in a glass sheet 110 scanned at locations of 1", 1.5", 2.0", 2.5" and 3" away from the sensor 112. The size of the defect does vary but the variation is small compared to other measurements and the variation is measurable and predictable. The graph also indicates that the calculated size of the defect gets bigger by a predictable amount as it is moved farther away from the sensor 112. Moreover, the graph indicates that the inverse of this size change which is the correction factor that one could use to calculate the correct size of the defect when the distance between the glass sheet 110 and the sensor 112 is known. The inspection system 100 also has several other advantages some of which are described below:

• The inspection system 100 has an optical geometry

(angle of incident light/angle of reflected light) that produces a pseudo or equivalent depth of field on the order of inches rather than millimeters. For large sheets of LCD glass (e.g., 2m by 2m) which cannot be handled or moved easily this means that the sensor can vary by inches relative to the glass surface and the imperfections can be still detected and measured.

• The inspection system 100 uses a transmission laser geometry that is a very efficient way to get light to the sensor 112. Other geometries that use small high-speed sensor elements are light starved which makes them impractical for the detection of 5 to 100 micron imperfections . • The inspection system 100 can also be implemented without the cylindrical lens 106 but the results would not be as accurate. For instance, in this alternative embodiment it would take more processing to calculate the presence and measurements of imperfections in the glass sheet 110.

Referring to FIGURES 2A-2C, there are shown three diagrams associated with a second embodiment of the inspection system 200 in accordance with the present invention. FIGURE 2A shows the inspection system 200 that includes a diode laser 202 which generates a laser line 204 that passes through a cylindrical lens 206 which refracts all of the rays in the laser line 204 into a beam of parallel light 208 that is transmitted at an angle which is close as possible to perpendicular through a glass sheet 210. During this process, part (about 4%) of the light beam 208 is reflected off the front surface of the glass sheet 210 and a part (about 4-5%) of the light beam 208 is reflected off the back surface of the glass sheet 210. The two reflected light beams 211 are received by the line-scan sensor 212. Since the laser beam 208 is highly coherent, the two reflected light beams 211 enable the sensor 212 and a computer (not shown) to generate an image of an interference pattern. As can be seen in FIGURE 2A, when a beam of coherent light 208 is directed at the glass sheet 210 there is also a reflection off both the front and back surfaces of the glass sheet 210 and the two waves produced go in and out of phase producing a fringe pattern (see FIGURE 2B) . This fringe pattern can be altered by either a change in the thickness or a change in the index of refraction of the glass sheet 210. The fringe patterns run in the north-south direction if there are small changes in the thickness and/or the index-of-refraction in the glass sheet 210. And, if there are a larger number of fringe streaks per unit area in the east-west direction then it indicates a more dramatic change in thickness and/or a more dramatic change in the index-of-refraction. As such, the coherent light 204 produced by the laser 202 allows the thickness or index of refraction change to be detected by the sensor 212 and mapped by a computer (not shown) . Moreover, the computer can be used to increase the precision of measuring the change in thickness or index- of-refraction of the glass sheet 210 by averaging the columns of the fringe patterns and finding the minimum and maximums of this sum (note the mins and the maxs are 90° out of phase and represent a change of thickness of ^XA the wavelength of light used to make the measurement) . One can also make measurements down to 1/10 the wavelength by subdividing the area between the min point and the max point into tenths .

Two exemplary interference patterns are shown in FIGURES 2B and 2C. In FIGURE 2B, each fringe in the interference pattern represents a change in the thickness of the glass sheet 210 equal to one-half the wavelength of the light beam 204. In FIGURE 2C, the line-scan-CCD generated image shows an inclusion in the glass sheet 210. It should be noted that the center portion of this image is a reflection off a 192 micron dark inclusion. This image also contains some fringes around the center portion. These fringes indicate a change in thickness and/or index-of-refraction that was caused by the dark inclusion within the glass plate 210.

It is possible to generate these images, because the two reflected light beams 211 form coherent waveforms which go in and out of phase (adding intensities and subtracting intensities) as they travel through space at an interval of % the wavelength of the laser light 204. For example, if a violet laser 202 was used with a wavelength of 400nm then one would see bright areas of the fringe each 200nm and dark areas of the fringe at 200nm intervals . The dark areas and the bright areas would be separated by approximately 1/6 of the wavelength of light 211. If violet light 211 is used then this bright to dark separation would be 66nm. It is because of this phenomena, that the sensor 212 can be used in a reflection bright field scanning (RBF) geometry so it can detect the bright and dark fringe pattern. And, by traversing this fringe pattern one can count the bright fringes (or dark fringes) and multiply this count by 1/3 of the wavelength of light 211 to determine the magnitude of thickness change in the glass sheet 210. In general, with this analysis itself it is not possible to determine if the fringe pattern change was caused by a thickness change or an index of refraction change. However, if one has experience in the glass making process they can analyze the interference patterns and determine what is causing the unique fringe pattern.

The inspection system 200 can also measure a distortion down to a percent of the wavelength of the laser light 204. This is possible since a fringe pattern is caused by the interference of two waveforms that go in and out of phase as they travel through space at an interval of 1/3 the wavelength of the laser light 204. One can associate the highest intensity (brightest portion) of the fringe with 0° and the lowest intensity

(darkest portion) can be associated with 90°. Then, one can infer where the point in the fringe pattern which is half way between the brightest and darkest portions would be which in this example is 45° because that corresponds to 1/12 of the wavelength of light 204. And, for a violet

400nm laser this would be approximately 30nm for violet light (404nm) . This is how one can interpret fringe patterns down to 1/12 a fringe.

The inspection system 200 also has several other advantages some of which are described below:

• If the angle of incident light 204 can be maintained relatively close to the normal with the glass sheet 210, then this optical geometry can produce a pseudo depth of field on the order of inches rather than millimeters. For large sheets of glass (e.g., 2m by 2m) which cannot be handled or moved easily it means that the sensor can vary by inches relative to the glass surface and the imperfections can be still detected and measured. The freedom gained with this scanning setup allows the glass sheet to be scanned as it is transported by the standard factory conveyance system.

• The inspection system 200 enables the measurement of sheet thickness as well as the detection of fine imperfections without needing to precisely position of the glass sheet. • The inspection system 200 generates local information as to whether an imperfection has distorted the sheet surface and this distortion can be measured down to a percent of the wavelength of the laser light used in the scanning.

• One can analyze the fringe pattern generated by the inspection system 200 and then determine a global change in thickness of the glass sheet

210.

• The inspection system 200 allows one to detect and measure a thickness or index of refraction change in the area of the inclusion. • The inspection system 200 can also detect any thickness or index of refraction change in the draw direction of the glass sheet 210 which would show itself as a streak going through the sight of the imperfection. • The inspection system 200 can also be implemented without the cylindrical lens 206 but the results would not be as accurate. For instance, in this alternative embodiment it would take more processing to calculate the thickness or index of refraction change in the glass sheet 210.

Referring to FIGURES 3A-3E, there are shown five diagrams associated with a third embodiment of the inspection system 300 in accordance with the present invention. FIGURE 3A shows the inspection system 300 which includes a sensor 302 and an illuminator 304 that are used to identify stress in a glass sheet 306. The illuminator 304 in this example includes a laser 306 and lens 308 (optional) that emit a polarized light beam 310a through a portion of the moving glass sheet 306. The sensor 302 (e.g., tri-linear sensor 302) uses three rows of detectors 312a, 312b and 312c (e.g., CCD detectors 312a, 312b and 312c) to receive the polarized light beam 310b that passed through the glass sheet 306 (see FIGURE 3B) . In this example, the polarized light beam 310a is 3- 5" wide. And, the sensor 302 is located approximately 2" away from the moving glass sheet 306. As shown in FIGURE 3B, the first row of CCD detectors 312a is blocked/covered with a first polarized coating 314a which polarizes incident light 310b with a zero degree orientation. The second row of CCD detectors 312b is blocked/covered with a second polarized coating 314b which polarizes incident light ^'31Ob with _. a 120° orientation relative to CCD detectors 312a. And, the third row of CCD detectors 312c is blocked/covered with a third polarized coating 314c which polarizes incident light 310b with a 240° orientation relative to CCD detectors 312a. Alternatively, it should be noted that the inspection system 300 can work with polarized coatings 312a, 312b and 312c can any three angles as long as the relative angular distances between them is around 120 degrees. The farther one varies form the relative angular distance of 120 degrees the less accurate the inspection system 300 becomes but it will still work. Angles such as 15 degrees, 135 degrees and 255 degrees will work as good as 0 degrees, 120 degrees, and 240 degrees because their relative angular distance is 120 degrees. Angles such as 15 degrees, 160 degrees, and 230 degrees will work but may not give the most accurate answer. As a result, the relative angular distance should be close to 120 degrees and any deviation from this ideal will result in a inspection system 300 which is less accurate but may still give an acceptable result.

In operation, when the sensor 302 is illuminated with the polarized light 310b, then the output from each row of CCD detectors 312a, 312b and 312c is a vector product of the input polarized light 310b and the polarizing filter angle associated with each row of CCD detectors 312a, 312b and 312c. Thus, when the polarized light 310a passes through a glass sheet 306 which contains a detectable amount of stress, then the stress changes the polarization angle of the light beam 310b which causes the signals from the three line-scan rows of CCD detectors 312a, 312b and 312c to also change relative to the amount of stress. These signals are used to identify stress in the glass sheet 306.

Thus, if there is no stress in the glass sheet 306, then the polarization angle of the received polarized light 310b will have the same angle as the light 310a emitted by the laser 306. And, if there is a small amount of stress present in the glass sheet 306, then this stress will alter the polarization angle of light 310b a small amount that can be measured and calculated by analyzing the output from the three rows of polarized CCD detectors 312a, 312b and 312c. And, if there is a large amount of stress in the glass sheet 306, then the angle of polarization of light 310b passing through the glass sheet 306 will be greatly altered and this change in the polarization again can be measured by the three rows of polarized CCD detectors 312a, 312b and 312c. One may think that it is possible to uniquely identify a polarization angle with two rows of CCD detectors 312a and 312b (for example) that contain orthogonal polarizers but there are cases on non-uniqueness . To illustrate this, reference is made to FIGURES 3C and 3D where two different polarization angles on two incoming waveforms translate into the same polarization magnitudes when they are projected onto two orthogonal polarization CCD detectors 312a and 312b. For these two waveforms it is impossible to uniquely identify their angle of polarization. This problem can be solved by adding a third row of CCD detectors 312c (for example) .

FIGURE 3E is a photo that shows an example of a piece of LCD glass 306 which is being dynamically flexed as a line-scan image is being generated by the sensor 302. The changes in fringe patterns over time indicate a change in stress. In general, the amount of stress that can vary over the area of the glass sheet 306 often depends on a number of environmental effects as well as how the glass sheet 306 was formed.

The inspection system 300 also has several other advantages some of which are described below:

• The inspection system 300 requires no moving parts .

• The inspection system 300 is suited for on-line measurement .

• The inspection system 300 can be used to generate a stress map of all areas on the LCD glass sheet 306. For instance, one can generate a complete stress map of the glass sheet 306, by using multiple sensors 302 aligned in a row to form, a long sensor as large as the glass sheet 306 and the signals generated from these sensors

302 can with the aid of a computer be used to generate a stress image of the entire glass sheet 306.

• As indicated above, the inspection system 300 can also be implemented without the cylindrical lens 308 but the results would not be as accurate. For instance, in this alternative embodiment it would take more processing to calculate/identify the stress in the glass sheet 306.

Referring to FIGURES 4A-4C, there are shown three diagrams associated with a fourth embodiment of the inspection system 400 in accordance with the present invention. FIGURE 4A shows the inspection system 400 which includes a color multi-line scan sensor 402 and multiple illuminators (lasers) 404a, 404b, 404c and 404d (four shown) that are used to identify imperfections within or on a glass sheet 406. In this example, the multi-line-scan sensor 402 has multiple rows of CCD detectors 412a, 412b, 412c and 412d each of which are covered by a spectral filter 414a, 414b, 414c and 414d (see FIGURE 4B) . And, the four different illuminators 404a, 404b, 404c and 404d each emit a color light beam 416a, 416b, 416c and 416d that has energy within an energy band of one of the filtered rows of CCD detectors 412a, 412b, 412c and 412d. FIGURES 4B-4C show how each spectral filter 414a, 414b, 414c and 414d allows only one particular color (wavelength) light beam 416a, 416b, 416c and 416d of interest to pass onto the corresponding row of CCD detectors 412a, 412b, 412c and 412d and blocks all other light beams 416a, 416b, 416c and 416d. In the exemplary inspection system 400 shown in FIGURES 4A-4C, the red illuminator 404a emits a red light beam 416a through a lens 418 and then through the glass sheet 406 onto the row of CCD detectors 412a filtered to receive the energy band of the red light beam 416a. In this example, the CCD detectors 412a are sensitive to microscopic inclusions in the glass sheet 406. The green illuminator 406b emits a green light beam 416b which is reflected off of the glass sheet 406 and directed into the row of CCD detectors 412b filtered to receive the energy- band of the green light beam 416b. In this example, the CCD detectors 412b are sensitive to inclusions and glass thickness. The blue illuminator 406c emits a blue light beam 416c through a grating 420 and then through the glass sheet 406 onto the row of CCD detectors 412c filtered to receive the energy band of the blue light beam 416c. In this example, the CCD detectors 412c are sensitive to streaks and changes in the index or refraction in the glass sheet 406. And, the grey (infra red) illuminator 406d emits a grey light beam 416d through a lens 424 and then through the glass sheet 406 onto the row of CCD detectors 412d filtered to receive the energy band of the grey (IR) light beam 406d. In this example, the CCD detectors 412d enables one to measure the position of an imperfection in the glass sheet 406. In a similar fashion, the inspection system 400 could be designed to detect other attributes in the glass sheet 406 using light beams in different energy bands like the infrared and ultraviolet energy bands. As can be seen, the inspection system 400 with one sensor 402 enables the measurement of a number of attributes about the contaminants and shape of the glass sheet 406.

For all practical purposes it does not matter which wavelength of light 416a, 416b, 416c and 416d is used to detect which type of attribute (e.g., microscopic inclusions, glass thickness). For instance, one could easily use the red light beam 416a and the CCD detectors 412a to detect changes in the index of refraction instead of microscopic inclusions in the glass sheet 406. Again, since the color spectral filters 414b, 414c and 414d are used to keep for example the red light 416a emitted from the red laser 404a from being seen by the non-red CCD detectors 412b, 412c and 412d. This means that the different colors of light 416a, 416b, 416c and 416d can be used to separate information provided by each attribute

(geometry- -angle of incident light and angle of reflected light on the sensor 402) from interfering with information generated by the other attributes (geometries) .

It also does not matter what the wavelengths of the four lasers 406a, 406b, 406c and 406d are so as long as they can be separated by spectral filters 414a, 414b, 414c and 414d which are placed in front of the four rows of the

CCD detectors 412a, 412b, 412c and 412d. Thus, the wavelengths of the lasers 406a, 406b, 406c and 406d can be selected to match inexpensive commercially available lasers such as 404nm, 750nm, 870nm, and 950nm. In addition, the wavelengths of light can be implemented at any useful wavelength from 200nm to 2000nm.

The inspection system 400 also has several other advantages some of which are described below:

• Spatial coordination: All of the measurements are generated from one multi-scan sensor 402 so it is much easier to coordinate a spatial relationship between the different views provided by each of the different line-scan-arrays 412a, 412b, 412c and 412d.

• Decreased cost : Here two or more rows of CCD detectors 412a, 412b, 412c and 412d can be implemented on one substrate which means one mounting device, one interface, and possibly one lens.

• Reduced size: In this case the inspection system 400 would have one sensor 402 instead of two or more which would take more space. Referring to FIGURES 5A- 5C, there are shown three diagrams associated with a fifth embodiment of the inspection system 500 in accordance with the present invention. It is well known today that inspection systems are used to scan a variety of materials (e.g., paper, plastics, steel, aluminum, and glass sheets) in order to detect and classify aberrations (imperfections) as they are being manufactured to gain quality control and process information. However, these scanning processes can be confused by foreign particles produced during the manufacturing process that land on the surfaces of the material and are detected by the inspection system. For transparent materials like a glass sheet, when a particle (e.g., dirt, dust, glass chips) is located on the surface of that material it can be equated to a particle (an inclusion) located inside the material by the inspection system. This leads to incorrect results produced by the inspection system. In fact, in some processes the number of surface particles may be 10 to 100 times the number of internal particles and this tends to make the scanning results meaningless. The inspection system 500 of the present invention addresses this problem by detecting imperfections 502 which are embedded in the transparent material 504 (e.g., glass sheet 504) without detecting surface particles 506.

As shown in FIGURES 5A and 5B, the inspection system 500 uses an illuminator 508 that emits light 510 towards the glass sheet 504 at a certain angle. The angle is chosen so that a part of the light 510 will be internally reflected within the moving glass sheet 504 to an area that is away from the position where the light 510 entered and exited the moving glass sheet 504. A line-scan camera 512 can then be placed in a position such that it can focus on this area and detect the light 510 reflected off the internal flaw 502 and not detect light 510 reflected off surface particles 506 that are located where the light 510 entered and exited the moving glass sheet 504. These two diagrams show the line-scan camera 512 that is located away from the point on the moving glass sheet 504 where light 510 from the illuminator 508 entered and exited the glass sheet 504. Again, the line-scan camera 512 in this position can focus on and detect the internal flaw 502 without detecting the surface particle 506.

In alternative embodiments, the line scan camera 512 can be replaced with line-scan sensors, time delay integration (TDI) sensors and contact sensors. And, the illuminator 508 can be a laser, a laser line, or any other illuminator such as a fluorescent light 508a (see FIGURE 5C) . If an illuminator like the fluorescent light 508a was used then a shield 514 would need to be used and positioned so as to allow light 510a to be internally reflected down the moving glass sheet 504 while at the same time block light 510a from entering or exiting the moving glass sheet 504 at a point where the line scan camera 512 is viewing the glass sheet 504 (see inspection system 500a shown in FIGURE 5C) . The inspection system 500 also has several other advantages one of which is described below:

• The inspection system 500 can be used to scan different product formats in addition to glass sheets 504 including for example glass webs, and other transparent materials in either plate or web format. Referring to FIGURES 6A- 6D, there are shown four diagrams associated with a sixth embodiment of the inspection system 600 in accordance with the present invention. It is well known to those skilled in the art of making glass that slight changes in the index of refraction and/or thickness of a glass sheet 602 that are created when the glass sheet 602 can refract a collimated light to a level where a difference in collimation is measurable. This effect can be detected by the human eye when looking at a glass sheet 602 (LCD display 602) and is considered an imperfection. FIGURES 6A and 6B illustrate this effect which is caused when light 604 is given off by a point source 606 (laser 606) and is transmitted through a flat undesirable glass sheet 602 that refracts the light 604 resulting in bright and dark streaks on a white background 608. The inspection system 600 of the present invention enables the detection of these slight variations in the thickness and/or index-of-refraction of a glass sheet 602 (or any transparent flat material) . This is important since undesirable glass sheets 602 can now be detected before they are used in a product like a LCD display.

FIGURE 6C shows the inspection system 600 which includes a laser 610 that creates a fan of light 612 of relatively equal light intensity. The inspection system 600 also includes a collimating lens 614 which diffracts the light 612 from a fan format into a line of parallel light 616. The light 616 is incident on a grating 618 which in this example has a period of 500 line pairs per inch and a 50% fill factor. The grating 618 forms a series of dark lines 622a and bright lines 622b which are projected through the glass sheet 602 (e.g., LCD glass sheet 602) onto a line-scan CCD sensor 620. In this example, the distance between the grating 618 and the glass sheet 602 is 2". The distance between the grating 618 and the sensor 620 is 4". And, the width of the beam of parallel light beam 616 is 3 "-5". If a piece of extremely flat "reference" glass sheet 602 with a constant thickness and index of refraction is analyzed by the inspection system 600, then a reference waveform like one shown as waveform 1 in the top of FIGURE 6D can be generated and stored in a computer (not shown) . Waveform 1 shows alternate bright and dark areas caused by the presence of the grating 618. The computer uses waveform 1 as a reference or standard to compare with waveforms from other glass sheets 602. For instance, if a relatively good glass sheet 602 is placed between the grating 618 and the sensor 620, then a waveform much like waveform 2 would be generated. If the tops are clipped off each waveform 1 and 2 so that the bright areas are equal and the two waveforms 1 and 2 are subtracted, then a waveform like waveform 3 would be generated. Waveform 3 shows small blips at the edges of the square wave which may be either positive or negative in magnitude. The width of the blips is relatively small because waveform 1 that is associated with the reference glass sheet 602 is approximately the same as the waveform 3 that is associated with the good glass sheet 602. Waveform 4 is an integral of the blips shown in waveform 3 which where produced by the positive edge of each waveform only (the blips produced by the negative edges are ignored) . In this case, the integral of the blips shown in waveform 4 is small since the good glass sheet 602 has almost the same quality as the reference glass sheet 602. More details about the relevance of the blips are described below when a waveform of a non-homogeneous glass sheet 602 is compared to the waveform of the reference glass sheet 602.

Waveform 6 was generated after the inspection of a non-homogeneous glass sheet 602 which had either a changing index of refraction or a thickness variation. The areas where the index of refraction was changing or where the thickness was not constant caused the direction of the light 616 to refract or change which then caused the edges of the waveforms to move from side to side. If the resulting waveform moved to the right, then the rays of light 616 that moved through the non-homogeneous glass sheet 602 were bent to the right. And, similarly if the waveform moved to the left then the rays of light 616 were bent to the left. If waveform 6 is subtracted from waveform 5 (which is the same as reference waveform 1) , then waveform 7 indicating the changing thickness, shape and/or index-of-refraction would be obtained. The "dark" shaded areas in waveform 7 are blips that are produced by the positive edges of waveform 5. And, the "light" shaded areas are produced by the negative edges of waveform 5. The width of the blips is an indication of the magnitude of the change in direction of light 616 as it traversed through the non-homogeneous glass sheet 602. The direction of the blips in waveform 7, either positive or negative, as compared to the edges of the waveform 5 determines the direction change in the light 616. For instance, the positive edges (the "dark" blips) of waveform 5 which are associated with positive blips indicate that the light rays were bent to the left. And, the negative "dark" blips in waveform 5 indicates that the light rays were bent to the right. If a line is drawn from the tops of all the "dark" blips, then waveform 8 is produced. If the integral value in waveform 8 is positive it indicates that the rays were bent to the left, if it is near zero then the rays were not bent, and if it is negative then the rays were bent to the right. Basically, the greater the value above zero in waveform 8 then the more undesirable the change in thickness, shape and/or index-of-refraction of the non-homogeneous glass sheet 602.

The inspection system 600 also has several other advantages one of which is described below:

• The inspection system 600 can be used to measure fine streaks in a glass sheet 602 which may be a result of a change in the index-of-refraction, a change in thickness or a change in the shape of the glass sheet 602. It generates both direction and relative magnitude information and it can generate a thousand readings per inch across the surface of the glass sheet 602.

Referring to FIGURES 7A- 7D, there are shown four diagrams associated with a seventh embodiment of the inspection system 700 in accordance with the present invention. Even though the aforementioned techniques used by the inspection systems 100, 200...600 to scan for microscopic imperfections are good at detecting imperfections in space due to the fact that they have a relatively large depth of field. However, this large depth of field (e.g., two inches) means that these scanning techniques do not have a good ability to detect the distance the imperfection is away from the sensor.

And, in the case of LCD glass it would be helpful if there was an inspection system that could determine if there is a defect and if that defect is on or near side A or side B of the LCD glass. This capability would be beneficial because during the process of coating LCD glass it becomes more sensitive to defects on one side than on the other side of the glass sheet. Therefore it is important to determine which side the defect is on because it may not be objectionable if it is on side B and very objectionable if it resides close to or is on side A. The inspection system 700 described below enables the determination of a position of a defect in the z direction which is the depth direction relative to the glass sheet.

FIGURE 7A shows a side view of an inspection system 700 using two laser line sources 702a and 702b each having different wavelengths and a sensor 706 with two line scan arrays 712a and 712b to determine the relative positions of two defects 706 and 708 which in this example are located at the same horizontal position on a glass sheet 710. The drawing shows the glass sheet 710 moving upward at a constant velocity (V) and at a fixed distance (D) from the sensor 704. The two line scan arrays 712a and 712b are a known distance apart (d) . And, each line- scan array 712a and 712b is sensitive to a different wavelength of light. For example, the bottom line-scan array 712b is sensitive to red light 714a emitted by a red laser 702a. And, the top line-scan array 712a is sensitive to green light 714b emitted by a green laser 702b. The top line- scan array 712a is illuminated at an angle of alpha (A) relative to the normal to the sensor 704. And, the bottom line-scan array 712b is illuminated by the bottom laser 702a at an angle normal to the glass sheet 710 and the sensor 704. In this example, the CCD line scan sensor 704 produces a new pixel every 5 microns and an image of both defects 706 and 708 is recorded in one or more scans. FIGURE 7A represents a snapshot of the inspection system 700 at time 0 when both defects 706 and 708 intercepted the light 714a from laser 702a.

FIGURE 7B is a snapshot of the inspection system 700 at time Tl when the defect 708 on side A of the glass sheet 710 intercepted the light 714b emitted from laser

702. The CCD line scan sensor 704 records the image of this defect 708 in one or more of its pixel scans.

FIGURE 7C is a snapshot of the inspection system 700 at time T2 when the defect 706 on side B of the glass sheet 710 intercepted the light 714b emitted from laser

702b. The CCD line scan sensor 704 records the image of this defect 706 in one or more of its pixel scans.

FIGURE 7D shows a composite of all the previous sensor scans and it shows three images of defects 706 and 708. First, it shows the image of both the A side defect

708 and the B side defect 706 superimposed over each other since they were detected at the same time 0. Second, it shows an image of the A side defect 708 as it passed through the light beam 714b emitted from laser 702b at time Tl. Third, it shows the image of the B side defect

706 at time T2 as it passed through light beam 714b emitted from laser 702b.

The distance the A side defect 708 has traveled can be calculated by counting the number of scan lines produced between time 0 and time Tl and then multiplying this number by 5 which is the size of a pixel in microns in this example. Likewise, the distance the B side defect 706 has traveled can be calculated by counting the number of scan lines produced between time 0 and time T2 and then multiplying this number by 5 which again in this example is the size of a pixel in microns. Because, it took the B side defect 706 longer to pass through light beam 714b than the A side defect 708 there will be more scan lines and a larger distance will be calculated for the B side defect 706.

After all of this, the distance that the A side defect 708 is from the sensor 704 can be calculated by multiplying the distance the A side defect 708 has traveled by the tangent of angle A. Likewise, the distance the B side defect 706 is from the sensor 704 can be calculated by multiplying the distance the B side defect 706 has traveled by the tangent of angle A. It should be appreciated that this type of distance calculation can be made for one or more defects that are located on the surface (s) of the glass sheet 710.

In an alternative embodiment, one can use the inspection system 700 to determine the position of a defect by monitoring its time of flight which is the time it takes the defect to move from the front of the first laser 702a until it intersects the second laser 702b.

Case I: The time of flight for the A side defect 708 can be calculated as follows:

t_a = (d + D* tan (A) ) /V

Case II: The time of flight for the B side defect 706 can be calculated as follows :

t_b = (d + (D+T(n_a/n_g) )*tan(A) ) /V

In the above equation n_a is the index of refraction of air and n_g is the index of refraction of glass sheet 710. Case III. The time of flight for a defect (not shown) that is positioned inside the glass sheet 710 at position P can be calculated as follows:

t_p = (d + (D+P(n_a/n_g) ) *tan(A) ) /V

If this equation is solved for the position P the equation becomes :

P = n_g* ( (t_p*V) -d-D*tan(A) ) / (n_a*tan(A) )

In an example, assume the angle of laser 702b is 20 degrees, the velocity of the glass sheet 710 is equal to 3 inches per second, the distance the glass sheet 710 is from the sensor 704 is 2 inches, the index of refraction for air to be 1, the distance between line- scan arrays 712a and 712b to be 90 microns, the pixel size is 5 microns, and the index of refraction of glass is 1.5. Then one can carry out the calculation at three different positions in the glass sheet 710: (1) at side A where P equals 0; (2) inside the glass where P equals 300 microns; and (3) at side B where P equals 700 microns. At these three locations the time of flight is as follows:

• At side B: time of flight = .243827 sec

• At 300 microns: time of flight = .243827 + .00095 = .244777 sec

• At side A: time of flight = .243827 + .00220 = .246027 sec

Then to calculate the positions of the defects, one needs to know the accuracy that the inspection system 700 can measure and in this example the sensor 702 can measure at 5 micron pixels. To achieve this resolution, a scan needs to be done at a rate equal to the velocity (V) of the glass sheet 710 divided by the 5 micron pixel size and this turns out to be 15240 scans per second. By inverting this number one gets the time between scans or 0.0000656 seconds per scan. This gives a scan difference of 33 between the time of flight generated from side B vs. side A for a 700 micron thick piece of glass sheet 710. As can be seen, this measurement is accurate enough to give good position information.

It should be noted that the sensors that can be used in any of the aforementioned inspection systems 100, 200...700 can be ones such as a Kodak KLI 14441 sensor and a Kodak KLI 4104 sensor. However, the specific type of sensor is not important. What is important is the fact that the sensor has multiple arrays of line-scan elements which helps in aligning and coordinating the information from the multiple camera views or multiple camera geometries . Although several embodiments of the present invention have been illustrated in the accompanying Drawings and described in the foregoing Detailed Description, it should be understood that the invention is not limited to the embodiments disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the spirit of the invention as set forth and defined by the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method for identifying imperfections in a transparent flat material, said method comprising the steps of: using an illuminator for emitting a light beam; using a first lens for receiving the light beam and emittirig a parallel light beam through a portion of the transparent -clat material; using a line-scan sensor for receiving the parallel light beam that passed through the portion of the transparent flat material and for focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

2. The method of Claim 1, wherein said parallel light beam that passed through the portion of the transparent flat material is structured by the imperfections and properties of the transparent flat material .

3. The method of Claim 1, wherein said line-scan sensor has a depth of field greater than one inch.

4. The method of Claim 1, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates an image that indicates a fresnel effect associated with the parallel light beam which passed through the transparent flat material that enables detection of the imperfections in the transparent flat material .

5. The method of Claim 1, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates an image that indicates an interference pattern associated with the parallel light beam which passed through the transparent flat material that enables detection of changes in thickness or index of refraction in the transparent flat material.

6. The method of Claim 1, wherein said transparent flat material is a glass sheet.

7. An inspection system for identifying imperfections in a transparent flat material, said inspection system comprising: an illuminator capable of emitting a light beam; a first lens capable of receiving the light beam and emitting a parallel light beam through a portion of the transparent flat material; a line-scan sensor capable of receiving the parallel light beam that passed through the portion of the transparent flat material and further capable of focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

8. The inspection system of Claim 1, wherein said parallel light beam that passed through the portion of the transparent flat material is structured by the imperfections and properties of the transparent flat material .

9. The inspection system of Claim 1, wherein said line-scan sensor has a pseudo depth of field greater than one inch.

10. The inspection system of Claim 7, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates an image that indicates a fresnel effect associated with the parallel light beam which passed through the transparent flat material that enables detection of the imperfections in the transparent flat material .

11. The inspection system of Claim 7, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates an image that indicates an interference pattern associated with the parallel light beam which passed through the transparent flat material that enables detection of changes in thickness or index of refraction in the transparent flat material. ^•

12. The method of Claim 7, wherein said transparent flat material is a glass sheet.

13. An inspection system for identifying imperfections in a transparent flat material, said inspection system comprising: an illuminator capable of emitting a light beam; a first lens capable of receiving the light beam and emitting a parallel light beam towards a portion of the transparent flat material; a line-scan sensor capable of receiving the parallel light beams that are reflected from a front surface and a back surface of the transparent flat material and further capable of focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

14. The inspection system of Claim 13, wherein said parallel light beams that are reflected from a front surface and a back surface of the transparent flat material are structured by the imperfections and properties of the transparent flat material.

15. The inspection system of Claim 13, wherein said line-scan sensor has a depth of field greater than one inch.

16. The inspection system of Claim 13, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates a reflection bright field fringe pattern image that enables detection of the imperfections in the transparent flat material.

17. The inspection system of Claim 13, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates a reflection bright field fringe pattern image that enables detection of changes in thickness or index of refraction in the transparent flat material .

18. The inspection system of Claim 13, wherein said transparent flat material is a glass sheet.

19. A method for identifying imperfections in a transparent flat material, said method comprising the steps of: using an illuminator for emitting a light beam; using a first lens for receiving the light beam and emitting a parallel light beam towards a portion of the transparent flat material; using a line-scan sensor for receiving the parallel light beams that are reflected from a front surface and a back surface of the transparent flat material and focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

20. The method of Claim 19, wherein said parallel light beams that are reflected from a front surface and a back surface of the transparent flat material are structured by the imperfections and properties of the transparent flat material.

21. The method of Claim 19, wherein said line-scan sensor has a depth of field greater ^vthan one inch.

22. The method of Claim 19, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates a reflection bright field fringe pattern image that enables detection of the imperfections in the transparent flat material.

23. The method of Claim 19, wherein said line-scan sensor outputs a signal that is analyzed by a computer which generates a reflection bright field fringe pattern image that enables detection of changes in thickness or index of refraction in the transparent flat material.

24. The method of Claim 19, wherein said transparent flat material is a glass sheet.

25. An inspection system for identifying stress in a transparent flat material, said inspection system comprising: a sensor; and an illuminator capable of emitting a polarized light beam through a portion of the transparent flat material towards said sensor; and said sensor capable of receiving the polarized light beam which passed through the transparent flat material by using: a first row of detectors that are blocked with a first polarizer; a second row of detectors that are blocked with a second polarizer; and a third row of detectors that are blocked with a third polarizer; and wherein if there is a detectable amount of stress present within the transparent flat material then that stress alters a polarization angle of the polarized light beam as it passed through the transparent flat material which can be detected and measured by the three rows of detectors .

26. The inspection system of Claim 25, wherein: said first polarizer is a 0° polarizer that coats said first row of detectors; said second polarizer is a 120° polarizer that coats said second row of detectors; and said third polarizer is a 240° polarizer that coats said third row of detectors.

27. The inspection system of Claim 25, further comprising a plurality of sensors which enable a stress map to be generated for all of the transparent flat material.

28. The inspection system of Claim 25, wherein said transparent flat material is a glass sheet.

29. A method for identifying stress in a transparent flat material, said method comprising the steps of: using an illuminator for emitting a polarized light beam through a portion of the transparent flat material towards a sensor; and using the sensor for receiving the polarized light beam which passed through the transparent flat material by using: a first row of detectors that are blocked with a first polarizer; a second row of detectors that are blocked with a second polarizer; and a third row of detectors that are blocked with a third polarizer; and wherein if there is a detectable amount of stress present within the transparent flat material then that stress alters a polarization angle of the polarized light beam as it passed through the transparent flat material which can be detected and measured by the three rows of detectors .

30. The method of Claim 29, wherein: said first polarizer is a 0° polarizer that coats said first row of detectors; said second polarizer is a 120° polarizer that coats said second row of detectors; and said third polarizer is a 240° polarizer that coats said third row of detectors.

31. The method of Claim 29, further comprising the step of using a plurality of the sensors which enable a stress map to be generated for all of the transparent flat material .

32. The method of Claim 29, wherein said transparent flat material is a glass sheet.

33. An inspection system for identifying different types of imperfections within or on a transparent flat material, said inspection system comprising: a sensor; a first illuminator capable of emitting a first light beam with a specific energy band that is reflected off from or passed through a portion of the transparent flat material towards said sensor; a second illuminator capable of emitting a second light beam with a specific energy band that is reflected off from or passed through a portion of the transparent flat material towards said sensor; said sensor capable of receiving the first and second light beams which reflected off or passed through the transparent flat material by using: a first row of detectors that are covered with a first spectral filter designed to permit only the first light beam to be received by the first row of detectors; and a second row of detectors that are covered with a second spectral filter designed to permit only the second light beam to be received by the second row of detectors; and wherein, each row of detectors enables detection of a different type of imperfection within or on the transparent flat material .

34. The inspection system of Claim 33, further comprising additional illuminators and additional rows of detectors and spectral filters.

35. The inspection system of Claim 33, wherein said imperfections include: a cord; a streak; a microscopic inclusion; and a change in a refraction index.

36. The inspection system of Claim 33, wherein said transparent flat material is a glass sheet.

37. A method for identifying different types of imperfections within or on a transparent flat material, said method comprising the steps of: using a first illuminator for emitting a first light beam with a specific energy band that is reflected off from or passed through a portion of the transparent flat material towards a sensor; using a second illuminator for emitting a second light beam with a specific energy band that is reflected off from or passed through a portion of the transparent flat material towards said sensor; using said sensor for receiving the first and second light beams which reflected off or passed through the transparent flat material by using: a first row of detectors that are covered with a first spectral filter designed to permit only the first light beam to be received by the first row of detectors; and a second row of detectors that are covered with a second spectral filter designed to permit only the second light beam to be received by the second row of detectors; and wherein, each row of detectors enables detection of a different type of imperfection within or on the transparent flat material.

38. The method of Claim 37, wherein additional illuminators and additional rows of detectors and spectral filters are used to enable the detection of even more types of imperfections within or on the transparent flat material

39. The method of Claim 37, wherein said imperfections include: a cord; a streak; a microscopic inclusion; and a change in a refraction index.

40. The method of Claim 37 , wherein said transparent flat material is a glass sheet.

41. An inspection system for identifying imperfections within a transparent flat material, said inspection system comprising: an illuminator capable of emitting a light beam at a predetermined angle towards the transparent flat material such that a portion of the light beam is internally reflected to an area in the transparent flat material that is located away from where the light beam originally entered and exited the transparent flat material; and a sensor capable of focusing on the area in the transparent flat material that is located away from where the light beam originally entered and exited the transparent flat material and further capable of detecting light reflected off imperfections within the transparent flat material without being effected by light reflected of particles on a surface of the transparent flat material.

42. The inspection system of Claim 41, wherein said illuminator includes at least one of the following: a laser; a laser line; and a light source in which a shield is used to allow light to be internally reflected within the transparent flat material while blocking light from entering and exiting the transparent flat material at a point where said sensor is focused on the transparent flat material.

43. The inspection system of Claim 41, wherein said sensor includes at least one of the following: a line-scan sensor; a line-scan camera; a time delay integration sensor; and a contact sensor.

44. The inspection system of Claim 41, wherein said transparent flat material is a glass sheet.

45. A method for identifying imperfections within a transparent flat material, said method comprising the steps of: using an illuminator for emitting a light beam at a predetermined angle towards the transparent flat material such that a portion of the light beam is internally reflected to an area in the transparent flat material that is located away from where the light beam originally entered and exited the transparent flat material; and using a sensor focused on the area in the transparent flat material that is located away from where the light beam originally entered and exited the transparent flat material to detect light reflected off imperfections within the transparent flat material without being effected by light reflected of particles on a surface of the transparent flat material.

46. The method of Claim 45, wherein said illuminator includes at least one of the following: a laser; a laser line; and a light source in which a shield is used to allow light to be internally reflected within the transparent flat material while blocking light from entering and exiting the transparent flat material at a point where said sensor is focused on the transparent flat material.

47. The method of Claim 45, wherein said sensor includes at least one of the following: a line-scan sensor; a line-scan camera; a time delay integration sensor; and a contact sensor.

48. The method of Claim 45, wherein said transparent flat material is a glass sheet.

49. An inspection system for detecting variations in thickness and/or index of refraction in a transparent flat material, said inspection system comprising: an illuminator capable of emitting a light beam; a lens capable of receiving the light beam and emitting a parallel light beam; a grating capable of receiving the emitted parallel light beam and then forming a series of bright lines and dark lines which are directed towards the transparent flat material; a sensor capable of receiving the series of bright lines and dark lines that passed through the transparent flat material and further capable of generating an output waveform related to the series of bright lines and dark lines that passed through the transparent flat material; and a processing unit capable of comparing the output waveform to a reference waveform and generating a difference waveform that indicates the variations in thickness and/or index of refraction of the transparent flat material.

50. The inspection system of Claim 49, wherein said difference waveform has at least one blip with a width that is an indication of a magnitude of a change in direction of one of the light beams as it passed through the transparent flat material.

51. The inspection system of Claim 49, wherein said transparent flat material is a Liquid Crystal Display (LCD) glass sheet.

52. A method for detecting variations in thickness and/or index of refraction in a transparent flat material, said method comprising the steps of: using an illuminator for emitting a light beam; using a lens for receiving the light beam and emitting a parallel light beam; using a grating for receiving the emitted parallel light beam and then forming a series of bright lines and dark lines which are directed towards the transparent flat material; using a sensor for receiving the series of bright lines and dark lines that passed through the transparent flat material and for generating an output waveform related to the series of bright lines and dark lines that passed through the transparent flat material; and using a processing unit for comparing the output waveform to a reference waveform and generating a difference waveform that indicates the variations in thickness and/or index of refraction of the transparent flat material.

53. The method of Claim 52, wherein said difference waveform has at least one blip with a width that is an indication of a magnitude of a change in direction of one of the light beams as it passed through the transparent flat material.

54. The method of Claim 52, wherein said transparent flat material is a Liquid Crystal Display (LCD) glass sheet.

55. An inspection system for identifying a position of a defect within or on a transparent flat material, said inspection system comprising: a sensor; a first illuminator capable of emitting a first light beam having a first wavelength that is passed through a portion of the transparent flat material towards said sensor; a second illuminator capable of emitting a second light beam having a second wavelength that is passed through a portion of the transparent flat material towards said sensor; said sensor including: a first row of detectors capable of receiving the first light beam which passed through the transparent flat material; a second row of detectors capable of receiving the second light beam which passed through the transparent flat material; and a processing unit capable of comparing images received from said first and second rows of detectors and then identifying the position of the defect within or on the transparent flat material.

56. The inspection system of Claim 55, wherein said processing unit compares the images received from said first and second rows of detectors and then identifies the position of the defect within or on the transparent flat material by: calculating a distance that the defect traveled which is within or on the transparent flat material that is moving between said sensor and said first and second illuminators; and multiplying the calculated distance by a tangent of an angle between the first light beam and the second light beam with respect to said sensor.

57. The inspection system of Claim 55, wherein: said first row of detectors is sensitive to the first light beam and not sensitive to the second light beam; and said second row of detectors is sensitive to the second light beam and not sensitive to the first light beam.

58. The inspection system of Claim 52, wherein said transparent flat material is a glass sheet.

59. A method for identifying a position of a defect within or on a transparent flat material, said method comprising the steps of: using a first illuminator for emitting a first light beam having a first wavelength that is passed through a portion of the transparent flat material towards a sensor; using a second illuminator for emitting a second light beam having a second wavelength that is passed through a portion of the transparent flat material towards said sensor which includes: a first row of detectors capable of receiving the first light beam which passed through the transparent flat material; a second row of detectors capable of receiving the second light beam which passed through the transparent flat material; and using a processing unit for comparing images received from said first and second rows of detectors and then identifying the position of the defect within or on the transparent flat material.

60. The method of Claim 59, wherein said processing unit compares the images received from said first and second rows of detectors and then identifies the position of the defect within or on the transparent flat material by: calculating a distance that the defect traveled which is within or on the transparent flat material that is moving between said sensor and said first and second illuminators; and multiplying the calculated distance by a tangent of an angle between the first light beam and the second light beam with respect to said sensor.

61. The method of Claim 59, wherein: said first row of detectors is sensitive to the first light beam and not sensitive to the second light beam; and said second row of detectors is sensitive to the second light beam and not sensitive to the first light beam.

62. The method of Claim 59, wherein said transparent flat material is a glass sheet.

63. An inspection system for identifying imperfections in a transparent flat material, said inspection system comprising: an illuminator capable of emitting a light beam through a portion of the transparent flat material; a line-scan sensor capable of receiving the light beam that passed through the portion of the transparent flat material and further capable of focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

64. A method for identifying imperfections in a transparent flat material, said method comprising the steps of: using an illuminator for emitting a light beam through a portion of the transparent flat material; using a line-scan sensor for receiving the light beam that passed through the portion of the transparent flat material and for focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

65. An inspection system for identifying imperfections in a transparent flat material, said inspection system comprising: an illuminator capable of emitting a light beam towards a portion of the transparent flat material; a line-scan sensor capable of receiving the light beams that are reflected from a front surface and a back surface of the transparent flat material and further capable of focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.

66. A method for identifying imperfections in a transparent flat material, said method comprising the steps of: using an illuminator for emitting a light beam towards a portion of the transparent flat material; using a line-scan sensor for receiving the light beams that are reflected from a front surface and a back surface of the transparent flat material and focusing on the imperfections in the transparent flat material without needing a second lens located in front of said line-scan sensor.