Automatic chip detection system and method

Abstract

The invention relates to the technical field of chip automatic detection, and discloses a chip automatic detection system and method, and the method comprises the steps: recording a pattern formed by bubbles in a chip under x-ray irradiation as a projection pattern; executing a circle domain planning strategy on the projection graph to obtain a circle where the projection graph is located, and recording the circle as an execution circle; obtaining execution circles corresponding to all the projection graphs in the execution rectangle to form an execution circle set; executing the pre-division strategy to obtain a plurality of execution units; aiming at any execution unit; obtaining the number of execution circles in the execution unit, and recording the number as a first side surface number; when the first side surface number is gt; 0, executing a three-dimensional detection strategy to obtain the volume of an execution circle on an execution section; and executing a qualification judgment strategy according to the threshold value proportion, and judging whether the chip is qualified or not. The number of bubbles in the chip is estimated by irradiating the three side surfaces of the chip with x light, and whether the chip is qualified or not is judged by calculating the bubble void ratio, so that the detection steps are simplified, and the detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of chip automatic detection, and in particular to a chip automatic detection system and method. Background Art

[0002] Chip automated testing is a key link in the semiconductor manufacturing process. It involves the use of computer-controlled equipment to test and control the quality of chips. With the development of integrated circuit technology, the complexity of chips has increased significantly, and the efficiency and accuracy of manual testing can no longer meet production needs. Therefore, automated testing technology has emerged. Automated testing technology usually includes visual inspection systems, electronic testing equipment, and dedicated software. The visual inspection system uses a high-resolution camera to capture images of the chip surface and identifies defects through image processing algorithms. Electronic testing equipment detects functional faults by applying electrical signals to the chip and measuring its response. These test data are then analyzed by dedicated software to determine whether the chip meets quality standards. Chip automated testing not only improves the speed and accuracy of detection, but also significantly reduces production costs while improving product reliability.

[0003] In the prior art, when detecting the bubble porosity of a chip to determine whether the chip is qualified, it is necessary to model the chip and analyze the bubbles inside the chip to determine whether the chip is qualified; modeling the chip is time-consuming and the detection efficiency is low.

[0004] This solution proposes to use X-rays to irradiate the three sides of the chip, obtain the number of execution circles photographed three times for each execution unit, calculate the number of execution circles on the execution section, and then obtain the volume of the bubbles in the area where the chip pad is located to determine whether the chip is qualified. Summary of the invention

[0005] The present invention provides a chip automatic detection system and method, which are used to promote solving the problems mentioned in the above background technology.

[0006] The present invention provides the following technical solution: a chip automatic detection method, comprising: The cuboid formed by the chip is referred to as the chip cuboid; In the chip cuboid: Any face of the chip cuboid is referred to as the first side face; Any surface perpendicular to the first side surface is referred to as the second side surface; Any one surface perpendicular to the first side surface and perpendicular to the second side surface is referred to as the third side surface; The chip to be tested is recorded as the chip to be tested; Irradiate the first side of the chip to be inspected with X-rays to obtain an inspection sample image; Obtain a rectangular area of ​​the pad of the chip to be detected on the detection sample image, which is recorded as an execution rectangle; The image formed by the bubbles inside the chip under X-ray irradiation is the projection image; Execute the circle domain planning strategy on the projected figure to obtain the circle where the projected figure is located, which is recorded as the execution circle; Get the execution circles corresponding to all the projected figures in the execution rectangle to form an execution circle set; The connecting line segment between the first side surface and the second side surface is recorded as the first edge; Setting a division interval for dividing the first edge; Execute the pre-division strategy to obtain multiple execution units; For any execution unit; Obtain the number of execution circles in the execution unit, recorded as the first side number; A plane perpendicular to the second side surface and parallel to the third side surface is drawn through the straight line passing through the execution unit, and the cross section of the chip cuboid cut by the plane is recorded as the execution cross section; When the number of the first side faces is greater than 0, the stereoscopic detection strategy is executed to obtain the volume of the execution circle on the execution section; Set the threshold ratio for determining whether the chip is qualified; According to the threshold ratio, the qualified judgment strategy is executed to determine whether the chip is qualified.

[0007] Optionally, executing a circular domain planning strategy on the projected graphics includes: Set the dividing line used to divide the projected figure; Set the segmentation distance used to determine the execution circle; Set the equidistant line, which is composed of multiple segmentation lines, and the distance between each line is the segmentation distance; Map the projected figure onto the isometric line; The line segment formed in the projection figure after the equidistant line intersects with the projection figure is recorded as the dividing line segment.

[0008] Optionally, the executing the circular domain planning strategy on the projected graphics further includes: For any segment: The split line segment divides the projected figure into two figures, which are respectively recorded as the first figure and the second figure; Set the differential value; Divide the segmented line segment into differential value line segments of equal length, and record the segment obtained by division as the bottom line segment; For any bottom edge segment: Draw a straight line perpendicular to the bottom line segment through the midpoint of the bottom line segment, and record it as the differential line; The point where the differential straight line intersects the first figure is recorded as the first marked point; The point where the differential straight line intersects the second figure is marked as the second marked point; Measure the distance between the first marked point and the bottom line segment, and record it as the first differential height; Measure the distance between the second marked point and the bottom line segment, and record it as the second differential height; Compute the area of ​​the first shape: Get the first differential height of each bottom edge segment, and record them as the first left differential height, the second left differential height, ... the eth left differential height; Get the length of the bottom line segment and record it as the bottom length; Execute the following formula: The first left differential height * base length + the second left differential height * base length + … + the eth left differential height * base length = the area of ​​the first figure.

[0009] Optionally, the executing the circular domain planning strategy on the projected graphics further includes: Compute the area of ​​the second shape: Get the second differential height of each bottom line segment, and record them as the first right differential height, the second right differential height, ... the eth right differential height; Execute the following formula: The first right differential height * base length + the second right differential height * base length + ... + the eth right differential height * base length = the area of ​​the second figure; Determine the size of the area of ​​the first figure and the area of ​​the second figure: If the area of ​​the first figure is equal to the area of ​​the second figure, obtain the midpoint of the segment and record it as the execution center; Get the first differential height and second differential height corresponding to each bottom edge segment on the segmentation segment; Compare all the heights and get the maximum height, which is recorded as the execution radius; The circle with the execution circle center as the center and the execution radius as the radius is recorded as the execution circle.

[0010] Optionally, the execution pre-division strategy is performed to obtain multiple execution units, including: Obtain two endpoints of the first edge, which are recorded as the first endpoint and the second endpoint respectively; Record the point that is one partition interval away from the first endpoint as the first execution point; Record the point 2 partition intervals away from the first endpoint as the second execution point; … The point m partition intervals away from the first endpoint is the mth execution point, wherein the mth execution point is the second endpoint; Draw a straight line perpendicular to the first edge through each execution point on the first side surface; The straight line and all the execution circles whose centers are on the straight line are collectively recorded as an execution unit.

[0011] Optionally, the executing a stereoscopic detection strategy includes: The execution circles in the execution unit are respectively the first execution circle, the second execution circle, ... the nth execution circle; Obtain the intersection of the straight line in the execution unit and the first edge, and draw a straight line perpendicular to the first edge through the intersection on the second side surface, which is recorded as a reference straight line; Irradiate the second side surface with X-rays to obtain execution circles with their centers on the control straight line to form a control execution circle set; Calculate the number of elements in the control execution circle set, and record it as the second side number; The connecting line segment between the first side surface and the third side surface is recorded as the second edge; On the first side surface, straight lines perpendicular to the second edge are drawn through the centers of the first execution circle, the second execution circle, ... the nth execution circle, respectively, and intersect the second edge at the first verification point, the second verification point, ... the nth verification point, respectively; Draw a straight line perpendicular to the second edge through each verification point on the third side to obtain a first verification straight line, a second verification straight line, and so on, an nth verification straight line; X-ray the third side: Obtain the number of execution circles whose centers are on the first verification line, which is recorded as the first verification number; Obtain the number of execution circles whose centers are on the second verification line, which is recorded as the second verification number; … Get the number of execution circles whose centers are on the nth verification line, recorded as the nth verification number; Get all the verification quantities to form a verification quantity set.

[0012] Optionally, the executing a stereoscopic detection strategy includes: Traverse all elements in the verification quantity set and compare them with the second side quantity; Get the elements that are less than the number of the second side, and record them as the first marked number, the second marked number, ... the vth marked number, where v≦n; Get the radius of each execution circle in the control execution circle set, calculate the mean of all radii, and record the result as the radius mean r; Calculate (nv)*number of second side faces+number of first marks+number of second marks+…+number of v marks=number of execution circles on the execution section, recorded as u; Calculate the volume k of all execution circles on the execution section, k= .

[0013] Optionally, the execution of the qualification determination strategy includes: Obtain the volumes of all execution circles on the execution cross sections corresponding to all execution units, calculate the sum of the volumes, and record the result as the bubble volume x; Get the volume y of the chip cuboid; Calculate x / y and record the result as the bubble porosity; When the bubble porosity is greater than or equal to the threshold ratio, the chip is unqualified; When the bubble porosity is less than the threshold ratio, the chip is qualified.

[0014] A system for an automated chip detection method, comprising: Data measurement module: measuring the distance between the first marking point and the bottom line segment; measuring the distance between the second marking point and the bottom line segment; measuring the radius of each execution circle in the comparison execution circle set; Data processing module: calculate the area of ​​the first figure; calculate the area of ​​the second figure; calculate the mean of all radii; calculate the number of execution circles on the execution section; calculate the volume of all execution circles on the execution section; calculate the volume of bubbles; calculate the porosity of bubbles; Data judgment module: judge the size of the area of ​​the first figure and the area of ​​the second figure to obtain the execution circle; compare the bubble porosity with the threshold ratio to judge whether the chip is qualified.

[0015] The present invention has the following beneficial effects: 1. The chip automatic detection method sets the segmentation line for segmenting the projection figure, sets the segmentation distance for determining the execution circle, sets the equidistant line, maps the projection figure onto the equidistant line, and records the line segments formed in the projection figure after the equidistant line intersects with the projection figure as segmentation line segments, which simplifies the processing of the figure, makes the calculation more intuitive and manageable, and improves the accuracy of the measurement, which is the key to subsequent calculations.

[0016] 2. The chip automated detection method sets a differential value, divides the segmented line segment into segments of equal length as the differential value, obtains a bottom line segment, measures the first differential height and the second differential height for each bottom line segment, divides the first figure into a plurality of rectangles with the bottom line segment as the width and the first differential height as the length, divides the second figure into a plurality of rectangles with the bottom line segment as the width and the second differential height as the width, calculates the area of ​​the rectangle, and then calculates the area of ​​the first figure and the area of ​​the second figure. When the area of ​​the first figure and the area of ​​the second figure are equal, obtains the execution center and the execution radius to obtain the execution circle, and confines the irregular figure to the circle, so as to facilitate the subsequent calculation of the bubble volume and improve efficiency.

[0017] 3. The chip automated detection method obtains the first endpoint and the second endpoint of the first edge, obtains all the execution points according to the divided intervals, draws a straight line perpendicular to the first edge through each execution point on the first side, and records the straight line and all execution circles whose centers are on the straight line as an execution unit, which helps to subdivide the detection tasks and improve the efficiency of management and analysis.

[0018] 4. The chip automatic detection method, when an execution circle appears on the first side, makes a reference line on the second side, irradiates the second side with X-rays, obtains the execution circle whose center is on the reference line, obtains the reference execution circle set, obtains the number of the second side, and the number of execution circles on the second side is the maximum number that appears directly below all the execution circles on the first side, which is used for the first estimation of all the execution circles on the execution section; it is convenient for the subsequent second estimation of the number of execution circles on the execution section.

[0019] 5. The chip automatic detection method is to draw a straight line perpendicular to the second edge through the center of each execution circle on the first side to obtain multiple verification points, and draw a straight line perpendicular to the second edge through each verification point on the third side to obtain multiple verification lines, and use x-rays to irradiate the third side to obtain the verification number on each verification line to form a verification number set. The number of execution circles on the execution section is estimated for the second time through the verification number set. When there is a verification number greater than the number on the second side, it means that when the third side is irradiated with x-rays, the verification number of execution circles on other execution sections interferes with this estimation, and it is excluded. When the verification number is less than the second side, it means that the verification number can more accurately calculate the number of execution circles on the execution section, and it is retained. The number of execution circles on the execution section is calculated by the second side number and the number of marks, and the number of execution circles on the execution section is more accurately estimated to improve the accuracy of detection.

[0020] 6. The chip automatic detection method uses the radius of each execution circle in the reference execution circle set to calculate the mean value, and obtains the radius mean as the radius of each execution circle on the execution section, and then calculates the volume of all execution circles on an execution section, obtains the volume of all execution circles on the execution section corresponding to all execution units, obtains the bubble volume, calculates the bubble porosity, and determines whether the chip is qualified by the threshold ratio and bubble porosity. Since there are more bubbles in the chip pad area and the bubbles are closer to a circle, only the bubbles in the pad area are calculated and other bubbles are excluded, which simplifies the calculation and improves efficiency.

[0021] 7. The chip automatic detection method uses X-rays to illuminate the three sides of the chip, calculates the bubble porosity, and determines whether the chip is qualified. If the chip is qualified, the step of chip modeling is omitted. If the chip is unqualified, the chip is modeled. The number of chips that need to be modeled is reduced, and the efficiency of detection is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the system of the present invention; Figure 2 is a schematic diagram of a test sample image of a first side surface irradiated with x-rays; Figure 3 Schematic diagram for executing rectangle and projected graphics; Figure 4 It is a schematic diagram of the relationship between the projection figure and the execution circle; Figure 5 is a schematic diagram of the positions of the first side surface, the second side surface and the third side surface in the cuboid; Figure 6 is a schematic diagram of a chip cuboid; Figure 7 A top view obtained by irradiating the first side surface with X-rays; Figure 8 is a left view obtained by irradiating the second side surface with X-rays; Fig. 9 is a front view obtained by irradiating the third side surface with X-rays; DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] In the first embodiment, the cuboid formed by the chip is referred to as a chip cuboid; Reference Figure 6 , schematic diagram of the chip rectangle.

[0025] In the chip cuboid: Any face of the chip cuboid is referred to as the first side face; Any surface perpendicular to the first side surface is referred to as the second side surface; Any one surface perpendicular to the first side surface and perpendicular to the second side surface is referred to as the third side surface; In this embodiment, the positions of the first side surface, the second side surface and the third side surface in the chip cuboid are shown in the figure.

[0026] The chip to be tested is recorded as the chip to be tested; Irradiate the first side of the chip to be inspected with X-rays to obtain an inspection sample image; In this embodiment, refer to Figure 2 , is the detection sample image; Obtain the rectangular area of ​​the pad of the chip to be detected on the detection sample image, which is recorded as the execution rectangle; The image formed by the bubbles inside the chip under X-ray irradiation is the projection image; In this embodiment, refer to Figure 3 , for executing rectangles and projected graphics; Execute the circle domain planning strategy on the projected figure to obtain the circle where the projected figure is located, which is recorded as the execution circle; Get the execution circles corresponding to all the projected figures in the execution rectangle to form an execution circle set; The connecting line segment between the first side surface and the second side surface is recorded as the first edge; Setting a division interval for dividing the first edge; Execute the pre-division strategy to obtain multiple execution units; For any execution unit; Obtain the number of execution circles in the execution unit, recorded as the first side number; A plane perpendicular to the second side surface and parallel to the third side surface is drawn through the straight line passing through the execution unit, and the cross section of the chip cuboid cut by the plane is recorded as the execution cross section; When the number of the first side faces is greater than 0, the stereoscopic detection strategy is executed to obtain the volume of the execution circle on the execution section; Set the threshold ratio for determining whether the chip is qualified; According to the threshold ratio, the qualified judgment strategy is executed to determine whether the chip is qualified.

[0027] The executing of the circular domain planning strategy on the projection graphics includes: Set the dividing line used to divide the projected figure; Set the segmentation distance used to determine the execution circle; Set the equidistant line, which is composed of multiple segmentation lines, and the distance between each line is the segmentation distance; Map the projected figure onto the isometric line; The line segment formed in the projection figure after the equidistant line intersects with the projection figure is recorded as the dividing line segment.

[0028] Set the dividing line used to divide the projection figure, set the dividing distance used to determine the execution circle, set the equidistant line, map the projection figure onto the equidistant line, and record the line segment formed in the projection figure after the equidistant line intersects with the projection figure as the dividing line segment. This simplifies the processing of the figure, makes the calculation more intuitive and manageable, and improves the accuracy of the measurement, which is the key to subsequent calculations.

[0029] The executing of the circular domain planning strategy on the projection graphics also includes: For any segment: The split line segment divides the projected figure into two figures, which are respectively recorded as the first figure and the second figure; Set the differential value; Divide the segmented line segment into differential value line segments of equal length, and record the segment obtained by division as the bottom line segment; For any bottom edge segment: Draw a straight line perpendicular to the bottom line segment through the midpoint of the bottom line segment, and record it as the differential line; The point where the differential straight line intersects the first figure is recorded as the first marked point; The point where the differential straight line intersects the second figure is marked as the second marked point; Measure the distance between the first marked point and the bottom line segment, and record it as the first differential height; Measure the distance between the second marked point and the bottom line segment, and record it as the second differential height; Compute the area of ​​the first shape: Get the first differential height of each bottom edge segment, and record them as the first left differential height, the second left differential height, ... the eth left differential height; Get the length of the bottom line segment and record it as the bottom length; Execute the following formula: The first left differential height * base length + the second left differential height * base length + … + the eth left differential height * base length = the area of ​​the first figure.

[0030] The executing of the circular domain planning strategy on the projection graphics also includes: Compute the area of ​​the second shape: Get the second differential height of each bottom line segment, and record them as the first right differential height, the second right differential height, ... the eth right differential height; Execute the following formula: The first right differential height * base length + the second right differential height * base length + ... + the eth right differential height * base length = the area of ​​the second figure; Determine the size of the area of ​​the first figure and the area of ​​the second figure: If the area of ​​the first figure is equal to the area of ​​the second figure, obtain the midpoint of the segment and record it as the execution center; Get the first differential height and second differential height corresponding to each bottom edge segment on the segmentation segment; Compare all the heights and get the maximum height, which is recorded as the execution radius; The circle with the execution circle center as the center and the execution radius as the radius is recorded as the execution circle.

[0031] In this embodiment, Figure 4 As shown, the relationship between the execution circle and the projected figure is shown.

[0032] Set the differential value, divide the segmented line segment into segments of equal length by the differential value, and obtain the bottom line segment. For each bottom line segment, measure the first differential height and the second differential height, divide the first figure into multiple rectangles with the bottom line segment as the width and the first differential height as the length, divide the second figure into multiple rectangles with the bottom line segment as the width and the second differential height as the width, calculate the area of ​​the rectangle, and then calculate the area of ​​the first figure and the area of ​​the second figure. When the area of ​​the first figure and the area of ​​the second figure are equal, obtain the execution center and execution radius to obtain the execution circle, and confine the irregular figure to the circle to facilitate the subsequent calculation of the bubble volume and improve efficiency.

[0033] The execution pre-division strategy is performed to obtain a plurality of execution units, including: Obtain two endpoints of the first edge, which are recorded as the first endpoint and the second endpoint respectively; Record the point that is one partition interval away from the first endpoint as the first execution point; Record the point 2 partition intervals away from the first endpoint as the second execution point; … The point m partition intervals away from the first endpoint is the mth execution point, wherein the mth execution point is the second endpoint; Draw a straight line perpendicular to the first edge through each execution point on the first side surface; The straight line and all the execution circles whose centers are on the straight line are collectively recorded as an execution unit.

[0034] Obtain the first endpoint and the second endpoint of the first edge, obtain all execution points according to the division interval, draw a straight line perpendicular to the first edge through each execution point on the first side, and record the straight line and all execution circles with their centers on the straight line as an execution unit. This helps to subdivide the detection tasks and improve the efficiency of management and analysis.

[0035] The implementation of the stereoscopic detection strategy includes: The execution circles in the execution unit are respectively the first execution circle, the second execution circle, ... the nth execution circle; Obtain the intersection of the straight line in the execution unit and the first edge, and draw a straight line perpendicular to the first edge through the intersection on the second side surface, which is recorded as a reference straight line; Irradiate the second side surface with X-rays to obtain execution circles with their centers on the control straight line to form a control execution circle set; Calculate the number of elements in the control execution circle set, and record it as the second side number; When an execution circle appears on the first side, a reference line is made on the second side, and the second side is irradiated with X-rays to obtain the execution circle with the center on the reference line, to obtain the reference execution circle set, and to obtain the number of execution circles on the second side. The number of execution circles on the second side is the maximum number that appears directly below all the execution circles on the first side, and is used for the first estimation of all execution circles on the execution section; this facilitates the subsequent second estimation of the number of execution circles on the execution section.

[0036] The connecting line segment between the first side surface and the third side surface is recorded as the second edge; On the first side surface, straight lines perpendicular to the second edge are drawn through the centers of the first execution circle, the second execution circle, ... the nth execution circle, respectively, and intersect the second edge at the first verification point, the second verification point, ... the nth verification point, respectively; Draw a straight line perpendicular to the second edge through each verification point on the third side to obtain a first verification straight line, a second verification straight line, and so on, an nth verification straight line; X-ray the third side: Obtain the number of execution circles whose centers are on the first verification line, which is recorded as the first verification number; Obtain the number of execution circles whose centers are on the second verification line, which is recorded as the second verification number; … Get the number of execution circles whose centers are on the nth verification line, recorded as the nth verification number; Get all the verification quantities to form a verification quantity set.

[0037] The implementation of the stereoscopic detection strategy includes: Traverse all elements in the verification quantity set and compare them with the second side quantity; Get the elements that are less than the number of the second side, and record them as the first marked number, the second marked number, ... the vth marked number, where v≦n; On the first side, a straight line perpendicular to the second edge is drawn through the center of each execution circle to obtain multiple verification points. On the third side, a straight line perpendicular to the second edge is drawn through each verification point to obtain multiple verification lines. The third side is irradiated with x-rays to obtain the verification number on each verification line to form a verification number set. The number of execution circles on the execution section is estimated for the second time through the verification number set. When there is a verification number greater than the number on the second side, it means that when the third side is irradiated with x-rays, the verification number of execution circles on other execution sections interferes with this estimation, and it is excluded. When the verification number is less than the second side, it means that the verification number can more accurately calculate the number of execution circles on the execution section, and it is retained. The number of execution circles on the execution section is calculated through the second side number and the number of marks, and the number of execution circles on the execution section is more accurately estimated to improve the accuracy of detection.

[0038] Get the radius of each execution circle in the control execution circle set, calculate the mean of all radii, and record the result as the radius mean r; Calculate (nv)*number of second side faces+number of first marks+number of second marks+…+number of v marks=number of execution circles on the execution section, recorded as u; In this embodiment, Figure 7 As shown, the first side is irradiated with X-rays, and the obtained figure is recorded as a top view, where n=3; The execution circles in the execution unit are respectively a first execution circle, a second execution circle and a third execution circle; like Figure 8 As shown, the second side is irradiated with X-rays, and the obtained figure is recorded as the left view, and the execution circle whose center is on the reference straight line is obtained to form a reference execution circle set; there are two execution circles in the reference execution circle set, which are recorded as the first reference circle and the second reference circle respectively; Get the second side quantity = 2; like Fig. 9 As shown, the third side is illuminated by X-ray, and the obtained figure is recorded as the front view; Among them, the first verification quantity on the first verification line = 2; the second verification quantity on the first verification line = 1; the third verification quantity on the third verification line = 3; traverse all elements in the verification quantity set and compare them with the second side quantity: the element smaller than the second side quantity is the second verification quantity, v = 1; The third verification quantity is obtained by mapping the execution circles on other execution sections onto the third verification straight line. Therefore, the third verification quantity is excluded.

[0039] Calculate (nv)*number of second sides+number of first marks+number of second marks+…+number of v marks=(3-1)*2+1=5=number of execution circles on the execution section, u=5; Calculate the volume k of all execution circles on the execution section, k= .

[0040] The radius of each execution circle in the reference execution circle set is used to calculate the mean value, and the radius mean is obtained as the radius of each execution circle on the execution section. The volume of all execution circles on an execution section is then calculated, and the volume of all execution circles on the execution sections corresponding to all execution units is obtained to obtain the bubble volume. The bubble porosity is calculated, and the qualification of the chip is determined by the threshold ratio and the bubble porosity. Since there are more bubbles in the chip pad area and the bubbles are closer to a circle, only the bubbles in the pad area are calculated and other bubbles are excluded to simplify the calculation and improve efficiency.

[0041] The execution of the qualified determination strategy includes: Obtain the volumes of all execution circles on the execution cross sections corresponding to all execution units, calculate the sum of the volumes, and record the result as the bubble volume x; Get the volume y of the chip cuboid; Calculate x / y and record the result as the bubble porosity; When the bubble porosity is greater than or equal to the threshold ratio, the chip is unqualified; When the bubble porosity is less than the threshold ratio, the chip is qualified.

[0042] Use X-rays to illuminate the three sides of the chip, calculate the bubble porosity, and determine whether the chip is qualified. If the chip is qualified, the chip modeling step is omitted. If the chip is unqualified, the chip is modeled. Reduce the number of chips that need to be modeled and improve the efficiency of detection.

[0043] Example 2, refer to Figure 1 , a system based on chip automated detection methods; Data measurement module: measuring the distance between the first marking point and the bottom line segment; measuring the distance between the second marking point and the bottom line segment; measuring the radius of each execution circle in the comparison execution circle set; Data processing module: calculate the area of ​​the first figure; calculate the area of ​​the second figure; calculate the mean of all radii; calculate the number of execution circles on the execution section; calculate the volume of all execution circles on the execution section; calculate the volume of bubbles; calculate the porosity of bubbles; Data judgment module: judge the size of the area of ​​the first figure and the area of ​​the second figure to obtain the execution circle; compare the bubble porosity with the threshold ratio to judge whether the chip is qualified.

[0044] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0045] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A chip automated detection method, comprising: The cuboid formed by the chip is referred to as the chip cuboid; In the chip cuboid: Any face of the chip cuboid is referred to as the first side face; Any surface perpendicular to the first side surface is referred to as the second side surface; Any one surface perpendicular to the first side surface and perpendicular to the second side surface is referred to as the third side surface; The chip to be tested is recorded as the chip to be tested; Irradiate the first side of the chip to be inspected with X-rays to obtain an inspection sample image; Obtain the rectangular area of ​​the pad of the chip to be detected on the detection sample image, which is recorded as the execution rectangle; The image formed by the bubbles inside the chip under X-ray irradiation is the projection image; Execute the circle domain planning strategy on the projected figure to obtain the circle where the projected figure is located, which is recorded as the execution circle; Get the execution circles corresponding to all the projected figures in the execution rectangle to form an execution circle set; The connecting line segment between the first side surface and the second side surface is recorded as the first edge; Setting a division interval for dividing the first edge; Execute the pre-division strategy to obtain multiple execution units; For any execution unit; Obtain the number of execution circles in the execution unit, recorded as the first side number; A plane perpendicular to the second side surface and parallel to the third side surface is drawn through the straight line passing through the execution unit, and the cross section of the chip cuboid cut by the plane is recorded as the execution cross section; When the number of the first side faces is greater than 0, the stereoscopic detection strategy is executed to obtain the volume of the execution circle on the execution section; Set the threshold ratio for determining whether the chip is qualified; According to the threshold ratio, the qualified judgment strategy is executed to determine whether the chip is qualified.

2. The chip automatic detection method according to claim 1, characterized in that: The executing of the circular domain planning strategy on the projection graphics includes: Set the dividing line used to divide the projected figure; Set the segmentation distance used to determine the execution circle; Set the equidistant line, which is composed of multiple segmentation lines, and the distance between each line is the segmentation distance; Map the projected figure onto the isometric line; The line segment formed in the projection figure after the equidistant line intersects with the projection figure is recorded as the dividing line segment.

3. The chip automatic detection method according to claim 2, characterized in that: The executing of the circular domain planning strategy on the projection graphics also includes: For any segment: The split line segment divides the projected figure into two figures, which are respectively recorded as the first figure and the second figure; Set the differential value; Divide the segmented line segment into differential value line segments of equal length, and record the segment obtained by division as the bottom line segment; For any bottom edge segment: Draw a straight line perpendicular to the bottom line segment through the midpoint of the bottom line segment, and record it as the differential line; The point where the differential straight line intersects the first figure is recorded as the first marked point; The point where the differential straight line intersects the second figure is marked as the second marked point; Measure the distance between the first marked point and the bottom line segment, and record it as the first differential height; Measure the distance between the second marked point and the bottom line segment, and record it as the second differential height; Compute the area of ​​the first shape: Get the first differential height of each bottom edge segment, and record them as the first left differential height, the second left differential height, ... the eth left differential height; Get the length of the bottom line segment and record it as the bottom length; Execute the following formula: The first left differential height * base length + the second left differential height * base length + … + the eth left differential height * base length = the area of ​​the first figure.

4. The chip automatic detection method according to claim 3, characterized in that: The executing of the circular domain planning strategy on the projection graphics also includes: Compute the area of ​​the second shape: Get the second differential height of each bottom line segment, and record them as the first right differential height, the second right differential height, ... the eth right differential height; Execute the following formula: The first right differential height * base length + the second right differential height * base length + ... + the eth right differential height * base length = the area of ​​the second figure; Determine the size of the area of ​​the first figure and the area of ​​the second figure: If the area of ​​the first figure is equal to the area of ​​the second figure, obtain the midpoint of the segment and record it as the execution center; Get the first differential height and second differential height corresponding to each bottom edge segment on the segmentation segment; Compare all the heights and get the maximum height, which is recorded as the execution radius; The circle with the execution circle center as the center and the execution radius as the radius is recorded as the execution circle.

5. The chip automatic detection method according to claim 1, characterized in that: The execution pre-division strategy is performed to obtain a plurality of execution units, including: Obtain two endpoints of the first edge, which are recorded as the first endpoint and the second endpoint respectively; Record the point that is one partition interval away from the first endpoint as the first execution point; Record the point 2 partition intervals away from the first endpoint as the second execution point; …… The point m partition intervals away from the first endpoint is the mth execution point, wherein the mth execution point is the second endpoint; Draw a straight line perpendicular to the first edge through each execution point on the first side surface; The straight line and all the execution circles whose centers are on the straight line are collectively recorded as an execution unit.

6. The chip automatic detection method according to claim 1, characterized in that: The implementation of the stereoscopic detection strategy includes: The execution circles in the execution unit are respectively the first execution circle, the second execution circle, ... the nth execution circle; Obtain the intersection of the straight line in the execution unit and the first edge, and draw a straight line perpendicular to the first edge through the intersection on the second side surface, which is recorded as a reference straight line; Irradiate the second side surface with X-rays to obtain execution circles with their centers on the control straight line to form a control execution circle set; Calculate the number of elements in the control execution circle set, and record it as the second side number; The connecting line segment between the first side surface and the third side surface is recorded as the second edge; On the first side surface, straight lines perpendicular to the second edge are drawn through the centers of the first execution circle, the second execution circle, ... the nth execution circle, respectively, and intersect the second edge at the first verification point, the second verification point, ... the nth verification point, respectively; Draw a straight line perpendicular to the second edge through each verification point on the third side to obtain a first verification straight line, a second verification straight line, and so on, an nth verification straight line; X-ray the third side: Obtain the number of execution circles whose centers are on the first verification line, which is recorded as the first verification number; Obtain the number of execution circles whose centers are on the second verification line, which is recorded as the second verification number; …… Get the number of execution circles whose centers are on the nth verification line, recorded as the nth verification number; Get all the verification quantities to form a verification quantity set.

7. The chip automatic detection method according to claim 6, characterized in that: The implementation of the stereoscopic detection strategy includes: Traverse all elements in the verification quantity set and compare them with the second side quantity; Get the elements that are less than the number of the second side, and record them as the first marked number, the second marked number, ... the vth marked number, where v≦n; Get the radius of each execution circle in the control execution circle set, calculate the mean of all radii, and record the result as the radius mean r; Calculate (nv)*number of second side faces+number of first marks+number of second marks+…+number of v marks=number of execution circles on the execution section, recorded as u; Calculate the volume k of all execution circles on the execution section, k= .

8. The chip automatic detection method according to claim 7, characterized in that: The execution of the qualified determination strategy includes: Obtain the volumes of all execution circles on the execution cross sections corresponding to all execution units, calculate the sum of the volumes, and record the result as the bubble volume x; Get the volume y of the chip cuboid; Calculate x / y and record the result as the bubble porosity; When the bubble porosity is greater than or equal to the threshold ratio, the chip is unqualified; When the bubble porosity is less than the threshold ratio, the chip is qualified.

9. A system for implementing the chip automatic detection method according to claim 1, comprising: Data measurement module: measuring the distance between the first marking point and the bottom line segment; measuring the distance between the second marking point and the bottom line segment; measuring the radius of each execution circle in the control execution circle set; Data processing module: calculate the area of ​​the first figure; calculate the area of ​​the second figure; calculate the mean of all radii; Calculate the number of execution circles on the execution section; calculate the volume of all execution circles on the execution section; calculate the volume of bubbles; calculate the porosity of bubbles; Data judgment module: judge the size of the area of ​​the first figure and the area of ​​the second figure to obtain the execution circle; compare the bubble porosity with the threshold ratio to judge whether the chip is qualified.

Images