TW201533438A - Focusing method for automated photo test equipment - Google Patents

Abstract

A focusing method is disclosed. The disclosed method comprise determines a plurality of candidate focusing length according to a focus length corresponding to a first device under test (DUT) among a plurality of DUTs, capturing a plurality of candidate focusing images corresponding to a second DUT next to the first DUT with the plurality of candidate focusing length, and selecting a candidate focusing image with the highest clarity value among the plurality of candidate focusing images as a base image of the second DUT to perform an examination process in advance.

Description

Focusing method of automatic optical detecting device

The present disclosure relates to a focusing method for an automatic optical detecting device, and more particularly to a method for an automatic optical detecting device to quickly focus on a plurality of objects to be tested.

In the field related to integrated circuits, automatic testing is an important part. The automatic test is a process for judging whether or not the function of an integrated circuit chip is normal. For an integrated circuit with an optical lens, it is necessary to judge whether or not the optical lens is flawed.

In determining whether an optical lens of an integrated circuit chip is flawed, an image of the optical lens can be captured, and the image is used to determine whether there is a scratch or a stain in the lens. To use images to determine if there are flaws in the lens, you must obtain a sufficiently clear image. Therefore, how to quickly focus on a large number of objects to be tested is a problem to be solved in optical automatic testing.

In view of the above problems, the present disclosure proposes a focusing method of an automatic optical detecting device, which focuses on one object or more adjacent to the object to be tested on the tray when focusing on one object to be tested on one tray Focus of the object to be tested Distance, to determine the focus distance of this object to be tested. Since the difference in focusing distance of adjacent objects to be tested on the tray is not large, this method can quickly focus on a plurality of objects to be tested, thereby reducing the time required for detection.

A focusing method of an automatic optical detecting device according to one or more embodiments of the present invention is adapted to focus on a plurality of device under test (DUT) on a tray, the focusing method comprising The focusing distance corresponding to the first object to be tested determines a plurality of candidate distances, and the first object to be tested is adjacent to the second object to be tested. Extracting a plurality of to-be-selected images corresponding to the second object to be tested by using the plurality of to-be-selected distances, and selecting, from the candidate images, the detection image of the second object to be tested with the highest definition for further imaging Test procedure.

Another focusing method of an automatic optical detecting device implemented according to one or more embodiments of the present invention is adapted to focus on a plurality of device under test (DUT) on a tray, the focusing method comprising according to the foregoing The focusing distance corresponding to the first object to be tested determines a first candidate distance and a second candidate distance of the second object to be tested, wherein the first object to be tested is adjacent to the second object to be tested. The first candidate image is captured on the second object to be tested by the first candidate distance. Taking a second candidate image for the second object to be tested by the second candidate distance. Comparing the sharpness of the first candidate image with the sharpness of the second candidate image to determine a relationship of the third candidate distance with respect to the first candidate distance or the second candidate distance. Determining a third candidate image for the second object to be tested by using the third candidate distance, and determining the detection image of the second object to be tested according to at least the first candidate image, the second candidate image, and the third candidate image To perform a detection procedure on the second object to be tested.

The focusing method of the automatic optical detecting device according to the present invention is based on The focus information of the object to be tested that is adjacent to the focus is determined to determine the possible focus distance of the current object to be tested, and the change trend of the focus distance may be further determined according to the plurality of objects to be tested, and may further be based on Two consecutive images of the same object to be tested determine the further focus direction, and the overall time required for focusing can be greatly reduced, thereby improving the efficiency of optical automatic testing.

The above description of the disclosure and the following description of the embodiments of the present invention are intended to illustrate and explain the spirit and principles of the invention, and to provide further explanation of the scope of the invention.

1‧‧‧Detection device

11‧‧‧ Storage Module

13‧‧‧Image capture module

15‧‧‧Processing module

17‧‧‧Test module

19‧‧‧Tray

20, 30, 40‧‧‧ Test objects

21, 31, 41‧‧ ‧ focus reference surface

23, 33, 43‧‧‧ parts to be tested

402‧‧‧Preset pattern

406‧‧‧瑕疵 image

f ₁ ~f ₁₃ ‧‧‧Selected focus plane

C ₂₀ , C ₃₀ ‧‧‧ sharpness distribution curve

d ₁ ‧‧‧displacement parameters

1 is a functional block diagram of a detecting device implemented in accordance with a method in an embodiment of the present invention.

Figure 2 is a schematic diagram of the operation of the detecting device implemented in accordance with the method in an embodiment of the present invention.

Figure 3 is a sharpness profile in accordance with an embodiment of the present invention.

Fig. 4A is an image showing the flawless portion of the object to be tested according to an embodiment of the present invention.

FIG. 4B is an image of a to-be-measured portion of the object to be tested according to an embodiment of the present invention.

Figure 5 is a schematic diagram of the operation of the detecting device implemented in accordance with the method of an embodiment of the present invention.

Figure 6 is a flow chart of a focusing method in accordance with an embodiment of the present invention.

Figure 7A is a partial flow chart of the focusing method in an embodiment of the present invention.

FIG. 7B is a partial flow chart of the focusing method in accordance with an embodiment of the present invention, which is continued from FIG. 7A.

FIG. 7C is a partial flow chart of the focusing method in accordance with an embodiment of the present invention, which is continued from FIG. 7A.

The detailed features and advantages of the present invention are set forth in the Detailed Description of the Detailed Description of the invention. The objects and advantages associated with the present invention can be readily understood by those skilled in the art. The following examples are intended to describe the present invention in further detail, but are not intended to limit the scope of the invention.

Referring to FIG. 1 and FIG. 2, FIG. 1 is a functional block diagram of a detecting device implemented by a method according to an embodiment of the present invention, and FIG. 2 is a second embodiment of the present invention. The figure is a schematic diagram of the operation of the detecting device implemented according to the method in an embodiment of the present invention. As shown in FIG. 1 , the detecting device 1 includes a storage module 11 , an image capturing module 13 , a processing module 15 , and a detecting module 17 . The processing module 15 is electrically connected to the storage module 11 and the image capturing module 13 . The detection module 17 is electrically connected to the processing module 15 . The detecting device 1 can be used to detect a plurality of objects to be tested on one tray.

The storage module 11 is configured to store a test object that has been in focus and/or detected by at least one of the plurality of test objects, for example, the object to be tested 20 in FIG. Corresponding focus distance, wherein the object to be tested 20 is located on one side (for example, the left side) of the object 30 to be in focus and/or detected, and as shown in FIG. 2, the object to be tested 20 may be adjacent to Measuring object 30. For example, when the detecting device 1 detects a plurality of objects to be tested in a row on a tray from left to right, when the detection of the object to be tested 20 is completed, the focus data of the object 20 to be tested, such as a focusing distance or It is a focus plane that can be stored in the storage module 11 as a basis for focusing on the object to be tested 30. In accordance with the spirit of the present invention, the storage module 11 can be a volatile storage medium or a non-volatile storage medium, which is not limited in the present invention.

In order to focus on one or more objects to be tested on the tray 19, the processing module 15 can generally control the image capturing module 13 to take multiple objects to be measured at a plurality of preset distances. Select an image. Taking FIG. 2 as an example, when the object 20 is first focused and detected, the processing module 15 (not shown in FIG. 2) controls the image capturing module 13 to focus at 13 preset distances. Go to the selected focus plane f ₁ to the selected focus plane f ₁₃ . Thus, thirteen images to be selected can be obtained. The depth of field of the image capturing module 13 can be selected to be equal to or slightly larger than the distance between two adjacent to-be-selected focusing planes (for example, the to-be-selected focusing plane f ₅ and the to-be-selected focusing plane f ₆ ). The post-processing module 15 analyzes the thirteen candidate images, each of which corresponds to a part of the image of the focus reference surface 21 of the object to be tested 20, and analyzes the partial image of each image to be selected to determine an image to be selected. "correctly" focuses on the focus reference surface 21 of the object to be tested 20.

Please refer further to FIG. 3, which is a sharpness profile in accordance with an embodiment of the present invention. Processing module 15 may be selected from these thirteen image block corresponding to the image focus of the reference surface 21 are recorded to the resolution, such as the sharpness of the distribution curve C in FIG. 3, ₂₀ is formed. Seen from FIG. 2, the reference surface 21 precisely aligned with the focus to be on the focal plane f ₇ is selected, it is possible by the distribution curve definition may be found in the corresponding C ₂₀ to be selected from the focal plane of the image f to be selected from the group _7, the in-focus reference surface 21 The image block has the highest resolution. Therefore, the processing module 15 selects the image to be selected corresponding to the to-be-selected focus plane f ₇ as the detected image of the object to be tested 20, and also uses the focus distance of the detected image. It is the preset distance that is focused on the to-be-selected focus plane f ₇ as a reference distance. The detection module 17 can then perform a detection procedure on the object to be tested 20 based on this (detection image and this reference distance).

Briefly, the processing module 15 analyzes the image blocks corresponding to the focus reference surface 21 for the thirteen candidate images to determine the clarity of the image blocks corresponding to the focus reference surface 21 among the thirteen candidate images. The image of the candidate image with the highest degree is selected as the detection image, and the corresponding focus distance is set as the reference distance, so that the detection module 17 performs the detection process on the object to be tested 20.

The detecting module 17 controls the image capturing module 13 to control the image capturing module 13 according to the aforementioned focusing distance (that is, focusing on the focusing reference surface 21 / to be selected for focusing in the detection program to be performed on the portion 23 to be tested of the object 20 to be tested. The focus distance of the plane f ₇ ), and the focus distance is reset according to a displacement parameter d ₁ to focus on the top end of the portion 23 to be tested. In the example of FIG. 2, the image capture module 13 is controlled to focus on the focal plane to be selected from the f _5, and can be sequentially selected to focus on the focal plane f _6, to be selected from the focal plane f _7, to be selected focal plane f ₈ , at the same time, the image to be tested 23 of the object 20 is subjected to one or more image captures to obtain one or more detection images. After the detected image is obtained, the detecting module 17 analyzes the detected image by optical identification to confirm whether the portion 23 to be tested has flaws.

For the method of optical identification, for example, please refer to FIG. 4A and FIG. 4B , and FIG. 4A is an image of the to-be-tested part of the object to be tested according to an embodiment of the present invention, and FIG. 4B is based on the present invention. In an embodiment of the invention, the portion to be tested of the object to be tested has a flawed image. Generally, if there is no flaw in the portion 23 to be tested, the detected image captured by the object to be tested will be as shown in FIG. 4A, and only the preset pattern 402 is included in the image block corresponding to the portion 23 to be tested in the image. . On the other hand, if there is flaws (scratches, impurities) in the portion 23 to be tested, as shown in FIG. 4B, in the image block corresponding to the portion 23 to be tested in the image, in addition to the preset pattern 402, there is an image of the image. 406. According to this, when the detecting module 17 determines that there is an unintended pattern in the image block corresponding to the portion to be tested in the detected image, it can be determined that the detected object to be tested has defects in the portion to be tested.

When the processing module 15 is to focus on the object to be tested 30, and the detection module 15 performs a detection process on the object to be tested 30, the processing module 15 controls the image capturing module 13 to repeat the foregoing object 20 for the object to be tested 30. A series of actions performed. However, when the number of objects to be tested is large, such a focus detection process is very time consuming.

In an embodiment of the present invention, returning to FIG. 2, after the processing module 15 performs the foregoing focusing process on the object to be tested 20, the focus reference surface 21 of the object to be tested 20 is recorded in the storage module 11. The corresponding focus distance will focus on the selected focus plane f ₇ . When the processing module 15 is to control the image capturing module 13 to focus on the object to be measured 30, the focus reference surface 31 of the object to be tested 30 and the focus reference surface 21 of the object to be tested 20 correspond to the to-be-selected focusing plane. Even if it is different, it will not be much worse (the vertical position of the two adjacent objects to be tested or the Z-axis position is substantially the same), so you can select N candidate focus planes centered on the to-be-selected focus plane f ₇ as Corresponding to the "possible focus plane" of the focus reference surface 31 of the object 30 to be tested, where N is a positive integer.

For example, N is 5, that is, a plurality of candidate focus planes are selected from the selected focus plane f ₅ to the selected focus plane f ₉ . Therefore, the processing module 15 can control the image capturing module 13 to correspond to the five selected distances of the five selected focus planes to be selected, respectively, to select the focus reference surface 31 of the object 30 to be selected. image. Thereby, the image capturing module 13 captures five images to be selected of the object to be tested 30, and the processing module 15 can separately analyze the sharpness of the five images to be selected, and obtain a sharpness distribution curve as shown in FIG. C ₃₀ . Finally, the processing module 15 can select one of the five candidate images corresponding to the image block corresponding to the focus reference surface 31 with the highest resolution. According to FIG. 3, the processing module 15 will be selected as the candidate to be selected to focus the image on the focal plane f _7, and the detecting device 30 as an image to be measured as in the previous treatment was carried out in a series of 20 Test procedure.

Please refer to FIG. 3 at the same time. It should be noted that even if the positional relationship between the object to be tested 20 and the object to be tested 30 (on the Z axis) is on the same horizontal plane as shown in FIG. 2, the object to be tested 20 is obtained. The sharpness distribution of the candidate image is still different from the sharpness distribution of the candidate image obtained by the object to be tested 30. Even if the image to be selected corresponding to the candidate focus plane (for example, the focus plane f ₇ to be selected) is the same, the sharpness is different. The cover factor is analyzed for the pattern on the focus reference surface of each object to be tested, such as the word "CAP" on Fig. 4A or Fig. 4B, but the pattern itself between each object to be tested The resolution will not be the same. Therefore, it is not possible to set a preset threshold value to simplify the process. Otherwise, the resolution of all the selected images corresponding to some objects to be tested may not exceed the preset. The threshold of sharpness is illegible, resulting in inability to focus.

The foregoing method directly focuses on the focus information of a previous object to be tested, such as a focus distance or a focus plane, to determine a plurality of focus distances or focus planes to be selected for capturing an image of the current object to be tested. For example, in the foregoing example, the focus plane f ₅ to be selected is correctly focused on the to-be-selected focus plane f ₇ of the focus reference surface 20, and the to-be-selected focus plane f _{9 is} selected as the image to be tested 30 for capturing the image to be selected. Multiple to-be-selected focus planes. However, it is possible to further speculate on the change trend of the focus distance by using the focus information of the plurality of test objects before, and appropriately select a plurality of possible focus planes or candidate distances according to the change trend, to the current The object to be tested captures a plurality of images to be selected. One of the implementations is exemplified below.

Please refer to FIG. 5, which is a schematic diagram of the operation of the detecting device implemented according to the method in an embodiment of the present invention. As shown in FIG. 5, when the processing module 15 is to focus on the object to be tested 20, the object to be tested 30, the object to be tested 40, and the object to be tested 50, first, the processing module 15 first controls the image capturing module. 13 aligning the object to be tested 20, respectively, focusing on the to-be-selected focus plane f ₁ to the to-be-selected focus plane f ₁₃ at a plurality of preset distances, and taking a total of thirteen images to be selected from the object to be tested 20, and then as shown in FIG. 5 As shown, the focus reference surface 21 is closest to the focus plane f ₆ to be selected, so that the focus distance of the object 20 to be inspected according to the foregoing method will be focused on the to-be-selected focus plane f ₆ , according to which the detection module 17 is detected by the aforementioned detection procedure. After the image capturing module 13 is focused on the to-be-selected focus plane f ₄ , the detected image is captured by the object 20 in a plurality of different focusing distances in the negative Z-axis direction to identify the detected portion 23 of the object 20 to be tested. defect.

Then, the processing module 15 controls the image capturing module 13 to be aligned with the object to be tested 30. At this time, according to the focus information of the object to be tested 20, that is, the focusing distance of the object to be tested 20 and the closest to the focusing reference surface 21 Select the focus plane f ₆ to select the focus plane f ₄ to be selected to the focus plane f ₈ to be selected. The processing module 15 controls the image capturing module 13 to focus on the five to-be-selected focusing planes to extract the image to be selected. According to the foregoing method flow, since the focus reference surface 31 of the object to be tested 30 is closest to the to-be-selected focus plane f ₅ as shown in FIG. ₅ , the processing module 15 can determine the candidate image corresponding to the to-be-selected focus plane f ₅ . It is a detection image of the object to be tested 30, so that the detection module 17 detects the portion 33 to be tested of the object 30 according to the foregoing detection procedure, and determines that a preset distance to focus on the to-be-selected focus plane f ₅ is to be The focusing distance of the object 30.

Then, when the processing module 15 controls the image capturing module 13 to align the object to be tested 40 to prepare the object 40 to be in focus. Since the focus information (focus distance or focus plane) of the object to be tested 20 and the object to be tested 30 have been recorded in the storage module 11, the focus reference surface 21 according to the object to be tested 20 is located substantially at the focus plane f ₆ to be selected. The focus reference surface 31 of the object to be tested 30 is substantially located on the information of the focus plane f ₅ to be selected, and the processing module 15 can predict that the focus reference surface 41 of the object to be tested 40 may be located substantially at the focus plane f ₄ to be selected. Therefore, the processing module 15 can control the image capturing module 13 to focus on the to-be-selected focusing plane f ₂ to the to-be-selected focusing plane f ₆ to capture the five images to be selected of the object 40 to be tested. It can be known from FIG. 5 that the focus reference surface 41 of the object 40 is closest to the focus plane f ₅ to be selected, so that the processing module 15 still calculates the image to be selected corresponding to the to-be-selected focus plane f ₅ , and focuses on The resolution of the image block of the reference surface 41 is higher than the definition of the same image block of the other several images to be selected. Therefore, the processing module 15 selects the image to be selected corresponding to the to-be-selected focus plane f ₅ as the detection image of the object 40 to be tested, and the detection module 17 detects the portion 43 to be tested of the object 40 by the aforementioned detection procedure.

It can be seen from the foregoing examples that, in an embodiment of the present invention, the processing module 15 does not simply determine the plurality of to-be-selected focus planes or multiple candidate distances of the object 40 to be tested by using the focus information of the object to be tested 30. . The processing module 15 can further predict and select a plurality of to-be-selected focus planes and/or to-be-selected distances of the object to be tested 40 according to the focus information of the object to be tested 20 and the focus information of the object to be tested 30.

Therefore, regarding the focusing method according to an embodiment of the present invention, reference may be made to FIG. 6, which is a flowchart of a focusing method according to an embodiment of the present invention. As shown in step S610, the image to be selected is respectively extracted from the first object to be tested by N preset distances, where N is a positive integer. As shown in step S620, the focus distance of the first object to be tested is determined according to the to-be-selected images of the N first objects to be tested. As shown in step S630, according to the focus distance of the first object to be tested, M adjacent candidate distances are selected from the N preset distances to align the second object to be tested adjacent to the first object to be tested. Take M candidate images, where M is less than N. As shown in step S640, the first image is determined according to the images of the M second objects to be tested. The focus distance of the two objects to be tested.

In another embodiment of the present invention, please continue to refer to FIG. 5, after the processing module 15 has controlled the image capturing module 13 to complete the focusing process of the object to be tested 20, the processing module 15 will focus on the object 20 to be tested. The information is recorded in the storage module 11, so that the processing module 15 controls the image capturing module 13 to focus on the object to be tested 30, and the processing module 15 can first control the image capturing module 13 to align the object to be tested 30, and Focusing on the to-be-selected focus plane f ₆ , the first candidate image of the object to be tested 30 is captured. The processing module 15 analyzes the sharpness of the image block corresponding to the portion of the focus reference surface 31 of the object 30 to be tested in the first candidate image.

Then, the processing module 15 can control the image capturing module 13 to focus on one of the to-be-selected focusing planes adjacent to the to-be-selected focusing plane f ₆ (for example, the in-focus plane f ₅ to be selected), and capture the object 30 to be tested. The second candidate image. The processing module 15 also analyzes the sharpness of the image block corresponding to the portion of the second reference image corresponding to the focus reference surface 31 of the object 30 to be tested, and the post-processing module 15 finds that it also corresponds to the focus reference surface 31. For some of the image blocks, the second image to be selected that focuses on the selected focus plane f ₅ has a higher definition. Accordingly, the processing module 15 can expect that if the image capturing module 13 is controlled to focus on the to-be-selected focusing plane f ₄ to capture the third candidate image, the third candidate image corresponds to the focusing reference surface. Part of the image block of 31 may have better definition.

When the processing module 15 actually controls the image capturing module 13 to focus on the to-be-selected focusing plane f ₄ to capture the third candidate image, and the processing module 15 actually analyzes the third candidate image, the processing module The group 15 finds that among the three candidate images, the image block corresponding to the portion of the focus reference surface 31 of the object 30 to be tested has the highest sharpness, so the processing module 15 can select the first The two candidate images are used as detection images of the object 30 to be tested. The detection module 17 then detects the image to be tested 33 of the object to be tested 30 by the aforementioned detection procedure according to the detection image and the corresponding focus information: focusing on the to-be-selected focus plane f ₅ .

Therefore, the focusing method in another embodiment of the present invention may include the processes of FIG. 7A to FIG. 7C, wherein FIG. 7A to FIG. 7C are respectively a part of the flow of the focusing method according to an embodiment of the present invention. Figure. As shown in step S710, the first candidate image is captured by the object to be measured at the first focus distance. As shown in step S720, the second candidate image is captured by the object to be measured at a second focus distance greater than the first focus distance. As shown in step S730, it is determined whether the resolution of the first candidate image is greater than the resolution of the second candidate image.

If the resolution of the first candidate image is greater than the resolution of the second candidate image, as shown in step S740, the third candidate image is captured by the object to be measured at a third focus distance smaller than the first focus distance. And as shown in step S742, it is determined whether the resolution of the third candidate image is smaller than the resolution of the first candidate image. If the resolution of the third candidate image is smaller than the resolution of the first candidate image, as shown in step S744, the first candidate image is used as the detection image of the object to be tested. Otherwise, as shown in step S746, the first focus distance is used as the new second focus distance, the third focus distance is used as the new first focus distance, and then the process returns to step S740.

If the resolution of the first candidate image is smaller than the resolution of the second candidate image, as shown in step S750: the third focus distance is greater than the second focus distance The third object to be selected is taken from the object to be measured. And as shown in step S752: determining whether the resolution of the third candidate image is smaller than the resolution of the second candidate image. If the resolution of the third candidate image is smaller than the resolution of the second candidate image, as shown in step S754, the second candidate image is used as the detection image of the object to be tested. Otherwise, as shown in step S756, the second focus distance is used as the new first focus distance, the third focus distance is used as the new second focus distance, and then the process returns to step S750.

In another embodiment, if the step S720 captures the second candidate image by using the second focus distance smaller than the first focus distance, then in step S730, when the resolution of the first candidate image is greater than Setting the third focus distance to be greater than the first focus distance when the brightness of the second candidate image is clear; setting the third focus distance to be smaller than the second focus when the resolution of the first candidate image is smaller than the resolution of the second candidate image Distance, the rest of the process is similar, no longer repeat them. Thus, with the spirit of the other parts of the present invention, even if only one object to be tested is focused three to four times, the correct focus for the object to be tested is completed, and a further detection procedure can be performed on the object to be tested.

According to the spirit of the present invention, the image capturing module 13 can apply a charge-coupled detector (CCD) or a complementary metal-oxide semiconductor photo detector (CMOS PD). Or any other device suitable for sensing and capturing images, and the image captured by the image capturing module 13 may be grayscale or full color image, which is not limited in the present invention.

Similarly, in accordance with the spirit of the present invention, the processing module 15 and the detection module 17 can be applied by a specific application integrated circuit (application-specific integrated Circuit, ASIC), advanced RISC machine (ARM), central processing unit (CPU), single-chip controller or other device suitable for executing arithmetic and control instructions, the present invention Not limited to this.

In summary, the focusing method and the detecting device using the method according to one or more embodiments of the present invention can be based on the focusing distance of adjacent objects to be tested when focusing on an object to be tested. Determining the possible focusing distance of the object to be tested, thereby greatly reducing the time for focusing on each object to be tested, and speeding up the process of automatic testing.

Although the present invention has been disclosed above in the foregoing embodiments, it is not intended to limit the invention. It is within the scope of the invention to be modified and modified without departing from the spirit and scope of the invention. Please refer to the attached patent application for the scope of protection defined by the present invention.

13‧‧‧Image capture module

19‧‧‧Tray

20, 30, 40‧‧‧ Test objects

21, 31, 41‧‧ ‧ focus reference surface

23, 33, 43‧‧‧ parts to be tested

f ₁ ~f ₁₃ ‧‧‧Selected focus plane

Claims (10)

A focusing method for an automatic optical detecting device for focusing a plurality of device under test (DUT) on a tray, the focusing method comprising: (a) a first waiting according to the objects to be tested The focusing distance of the measuring object determines M candidate distances, the M candidate distances are different from each other, and M is an integer greater than one; (b) extracting M pieces from the second candidate object by the M candidate distances a candidate image to be selected, the second object to be tested is adjacent to the first object to be tested, and the candidate images of the M second objects to be tested respectively correspond to the M candidate distances; and (c The image with the highest resolution among the candidate images of the M second analytes is selected as a detection image of the second object to be tested. The focusing method of the automatic optical detecting device according to claim 1, wherein in the step (a), the focusing distance of the first object to be tested is taken as a center to determine the M candidate distances. The focusing method of the automatic optical detecting device of claim 1, further comprising: (d) extracting, to the first object to be tested, N candidate images of the first object to be tested by N preset distances, the N The to-be-selected images of the first object to be tested respectively correspond to the N preset distances, N is an integer greater than M; and (e) the image with the highest resolution among the to-be-selected images of the N first objects to be tested , determining the focusing distance of the first object to be tested. A focusing method for an automatic optical detecting device for focusing a plurality of device under test (DUT) on a tray, the focusing method comprising: (a) a first waiting according to the objects to be tested The focus distance of the object is determined by one a first focus distance and a second focus distance; (b) capturing a first candidate image for the second object to be tested by the first focus distance, the second object to be tested being adjacent to the first object to be tested (c) capturing a second candidate image for the second object to be tested by the second focus distance; (d) determining a resolution of the first image to be selected and a brightness of the second image to be selected Determining a third focus distance; (e) drawing a third candidate image for the second object to be measured by the third focus distance; and (f) when the third candidate image has a lower definition than the first When a candidate image or the second candidate image is selected, a detected image of the second object to be tested is determined according to the sharpness of the first candidate image and the sharpness of the second candidate image. The focusing method of the automatic optical detecting device of claim 4, further comprising: (g) when the resolution of the third candidate image is greater than the first candidate image and the second candidate image, according to the first The three focus distance determines the new first focus distance and the second focus distance, and repeats steps (b) through (e). The focusing method of the automatic optical detecting device of claim 4, wherein in step (a), the focusing distance of the first object to be tested is used as the first focusing distance, and the second focusing distance is greater than the first A focus distance. The focusing method of the automatic optical detecting device of claim 6, wherein in the step (d), the method includes: (d1) when the resolution of the first candidate image is greater than the second candidate image In the case of sharpness, the third focusing distance is set to be smaller than the first focusing distance; and (d2) when the resolution of the first candidate image is smaller than the resolution of the second candidate image, setting the third focusing distance to be greater than The second focus distance. The focusing method of the automatic optical detecting device according to claim 4, wherein in step (a), the focusing distance of the first object to be tested is used as the first focusing distance, and the second focusing distance is smaller than the first A focus distance. The focusing method of the automatic optical detecting device of claim 8, wherein in the step (d), the method includes: (d1) when the sharpness of the first candidate image is greater than the sharpness of the second candidate image, Setting the third focus distance to be greater than the first focus distance; and (d2) setting the third focus distance to be smaller than the second focus when the resolution of the first candidate image is smaller than the resolution of the second candidate image distance. The focusing method of the automatic optical detecting device of claim 4, further comprising: (h) extracting, from the plurality of predetermined objects, a plurality of to-be-selected images of the first object to be tested, the plurality The plurality of to-be-selected images of the first object to be tested respectively correspond to the plurality of preset distances; and (i) determining the first to-be-tested according to the image with the highest resolution among the plurality of to-be-selected images of the plurality of first objects to be tested The focus distance of the object.