A method to assist R&D personnel in analyzing the problem of low probability of not booting

Through image acquisition and digital image processing technology, the startup process of industrial computers is monitored, and the failed videos are automatically detected and saved, which solves the monitoring problem of low-probability failure, improves detection efficiency and accuracy, and ensures that the product is on the market on time.

CN112416689BActive Publication Date: 2024-11-12ENNOCONN SUZHOU TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202011414309.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-03
Publication Date
2024-11-12
Estimated Expiration
2040-12-03

AI Technical Summary

Technical Problem

The existing technology is difficult to efficiently monitor and analyze the problem of low-probability failure, which leads to waste of R&D resources and prolongs debug cycles, which may be missed by human negligence, which affects the product's launch as scheduled.

Method used

The image acquisition device collects the boot process video of the device to be tested, and uses the digital image processing method to detect whether the boot process screen meets the specifications. If it does not meet, the failed video will be saved, and the test will be automatically stopped through the counting and alarm mechanisms.

Benefits of technology

Effectively save manpower, shorten the time for analyzing problems, reduce the risk of human detection errors, improve detection efficiency, ensure that products are on time or in advance, simplify testing work, realize automation and intelligence, and reduce the total project cost.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112416689B_ABST
    Figure CN112416689B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for assisting R&D personnel in analyzing a problem of a small probability of not starting up, comprising the following steps: connecting a device to be tested to a power source and powering on; controlling the on and off of a power socket according to a preset power-on holding time, a countdown, and the number of power-on and power-off tests, so that the device to be tested is in a power-on or power-off state; after the device to be tested switches from a power-off state to a power-on state each time, collecting a power-on process video of the device to be tested by an image acquisition device; extracting a power-on process screen from the power-on process video, detecting the power-on process screen by a digital image processing method, and if it meets the detection specifications, saving the detection results and continuing the next power-on and power-off detection; if it does not meet the detection specifications, saving the power-on process failure video of the device to be tested. The present invention has fast detection speed, low error rate, can save manpower, and shortens the time for analyzing problems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment power on / off detection, and in particular to a method for assisting R&D personnel in analyzing a small probability power-off problem. Background Art

[0002] Industrial computers require very high stability, so powering on and off more than 1,000 times is a must-test item, and not even a single failure is allowed.

[0003] In the past, when encountering a low-probability failure phenomenon that was very difficult to reproduce, R&D personnel had to stay next to the test motherboard to monitor the test until they saw the failure phenomenon occur. This testing method seriously wastes R&D resources and may miss the failure phenomenon due to human negligence of R&D personnel, resulting in a very long debug cycle, which will seriously affect the scheduled launch and sales of the product. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method to assist R&D personnel in analyzing the problem of a small probability of not starting up. In the system and method,

[0005] In order to achieve the above object, the technical solution of the present invention is as follows:

[0006] A method for assisting R&D personnel in analyzing a small probability non-startup problem includes the following steps:

[0007] The device under test is connected to the power supply and powered on;

[0008] According to the preset power-on holding time, countdown and power-on and power-off test times, the power socket is controlled to be turned on or off, so that the device to be tested is in the power-on or power-off state;

[0009] After the device under test is switched from a shutdown state to a startup state each time, a startup process video of the device under test is captured by an image capture device;

[0010] The boot process screen is extracted from the boot process video, and the boot process screen is detected by digital image processing method. If it meets the detection specifications, the detection result is saved and the next power on / off detection is continued; if it does not meet the detection specifications, the failed boot process video of the device to be tested is saved.

[0011] Preferably, the method further includes the steps of counting the on / off state of the device under test, and when the device under test switches from the off state to the on state, the count is increased by 1; when the count number is equal to the preset on / off test number, the on / off test is stopped.

[0012] Preferably, the boot process screen includes a boot initial stage screen, a screen before entering the operating system, a screen for booting into the operating system, and a screen for successfully entering the operating system.

[0013] Preferably, the method further includes the following steps: detecting the boot process screen by a digital image processing method, and issuing a prompt alarm if the screen does not meet the detection specifications.

[0014] Preferably, a digital image processing method is used to detect whether the boot process screen meets the specifications, and the specific steps are as follows:

[0015] Extracting image blocks with indicative icons from the boot process screen;

[0016] Cut the image block according to the standard template to obtain m regional images;

[0017] The m regional images are compared and matched with the corresponding regional images in the sample image in turn for similarity. If the comparison results of the m regional images are all less than the preset threshold, the similarity comparison and matching of the next image block is performed until the comparison results of all image blocks are less than the preset threshold, then it is determined that the boot process screen meets the specifications; otherwise, it is determined that the boot process screen does not meet the detection specifications.

[0018] Preferably, the similarity comparison and matching adopts a similarity matching method of hash coding and color histogram.

[0019] Preferably, the boot process screen is pre-processed.

[0020] Preferably, the preprocessing is median filtering.

[0021] Based on the above technical solution, the beneficial effects of the present invention are:

[0022] 1) In the field of electronics and industrial products, if high-frequency and repetitive testing is required to reproduce the problem, this device can be used for monitoring, thereby effectively saving manpower and shortening the time for analyzing the problem;

[0023] 2) Saving the failed startup process video of the device under test, so that the R&D personnel can deduce the failure cause based on the failure video, thus saving the troubleshooting time of the R&D personnel;

[0024] 3) Can effectively block various undesirable phenomena;

[0025] 4) Improve the time and efficiency of testing, reduce the risk of human testing errors, and help products be launched on time or ahead of schedule;

[0026] 5) Integrate software and hardware design. This device is used to monitor the test process of the device under test and save test results and test logs;

[0027] 6) Solve industry pain points, simplify the complexity of testing staff, realize automation and intelligence of testing work, thereby reducing the overall project cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0029] Figure 1 : A flow chart of a method for assisting R&D personnel in analyzing a small probability of a computer not starting up;

[0030] Figure 2 : The present invention provides a method for assisting R&D personnel in analyzing a small probability of a computer not starting up, and the analysis interface diagram a;

[0031] Figure 3 : Analysis interface diagram b in a method for assisting R&D personnel in analyzing a small probability of a computer not starting up problem according to the present invention;

[0032] Figure 4 :The present invention is an example diagram of the position of the selection box in the standard template in different rounds, wherein (a) is an example diagram of the position of the selection box in the standard template of the count round; (b) is an example diagram of the position of the selection box in the standard template of the count+1 round. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0034] Embodiment 1

[0035] like Figures 1 to 4 As shown, the present invention provides a method for assisting R&D personnel in analyzing a small probability of a computer not starting up, comprising the following steps:

[0036] Connect the industrial computer to the power supply and turn it on;

[0037] According to the preset power-on time, countdown and power-on and power-off test times, the power socket is controlled to be turned on or off to turn the industrial computer on or off;

[0038] Each time the industrial computer switches from the shutdown state to the startup state, the startup process video of the device under test is collected through the camera;

[0039] The boot process screen is extracted from the boot process video, and the boot process screen is detected by digital image processing methods. If it meets the detection specifications, the detection result is saved and the next power on / off detection is continued; if it does not meet the detection specifications, the failed boot process video of the industrial computer is saved and a prompt alarm is issued.

[0040] The boot process screen includes the initial boot stage screen, the screen before entering the operating system, the screen of booting into the operating system, and the screen of successfully entering the operating system. The four stages of the entire boot process must be tested in turn to be qualified before the test results can be saved and the next power on and off test can be continued.

[0041] Furthermore, the method also includes the following steps: counting the on / off state of the industrial computer, when the industrial computer switches from the off state to the on state, the count is increased by 1; when the count number is equal to the preset on / off test number, the on / off test is stopped.

[0042] Furthermore, the method of digital image processing is used to detect whether the boot process screen meets the specifications. Taking the test of the initial stage screen of the boot process screen of a 300*300 as an example, the specific steps are as follows:

[0043] Step 0: If there is no sample image, first start the industrial computer, manually perform a normal power on and off, determine the position of the indicative icon in the initial stage of the boot screen, and use the camera to capture the image block with the indicative icon as the sample image, recorded as I model , and save; if you already have relevant sample images, you can start directly from Step 1;

[0044] Step 1: Load the boot process sample image I model , cut the sample image according to the requirements of the standard template to obtain 4 regional images I m1 ...I m4 ,The properties of the selection box in the standard template are shown in Table 1. At the same time, the four region images can be respectively median filtered, the color histogram and the hash coding operations can be calculated;

[0045] Step 2: Set the timer, turn on the camera and start matching. The camera captures the image block with the indicative icon in the initial stage of the industrial computer startup process in real time according to the image size of 300*300, which is recorded as captured image I;

[0046] Step 3: According to the standard template in Step 1, the image block I with the indicative icon captured in Step 2 is cut to obtain four regional images I1...I4. The cutting positions are the same as the sample images.

[0047] Table 1 Selection box attributes in standard templates

[0048] template Marquee 1 Marquee 2 Marquee 3 Marquee 4 size 100*100 75*75 75*75 60*60

[0049] Step 4: Perform median filtering on the region images I1 ... I4 of the cut captured image I to reduce image noise and thereby reduce the impact of noise on the test results.

[0050] Step 5: Calculate the color histogram and hash code of the captured image area images I1...I4 after denoising in Step 4.

[0051] Step 6: Compare the color histogram and hash code of the captured images I1...I4 obtained in Step 5 with the corresponding area image I in the sample image m1 ...I m4 The color histogram and hash code of the four selection boxes are matched and compared. When the color histogram and hash code in the four selection boxes are less than the corresponding preset threshold, it can be considered that the captured image I and the sample image I model Similar, otherwise, it is considered that the sample image I model It is not similar to the acquired image I taken by the camera.

[0052] Further, the method steps of using color histogram for similarity matching are as follows:

[0053] 1) Select the box from the sample image I model The obtained subgraph is G j (g1, g2, ..., g m ), where j∈1, 2, 3, 4 represent the selection box number, m represents the total number of pixels in the image, G j Indicates that the selection box j is from I model The image obtained, g i Represents the pixel value at position i in the image, i∈1,2,...,m; capture the selection box of image I to obtain the sub-image S j (s1,s2,...,s m ) indicates that j and m have the same meaning as above, S j represents the image obtained by the selection box j from I, s i Represents the pixel value at position i in the image, i∈1, 2, ..., m;

[0054] 2) Calculate the Euclidean distance of the corresponding selection box image. The Euclidean distance formula is:

[0055]

[0056] ED j The closer the value of (G, S) is to 0, the more similar the two marquee images are. j (G j , S j) meets the requirements (i.e., is smaller than a preset appropriate threshold), the two images can be judged to be similar.

[0057] Further, the method steps of using hash coding for similarity matching are as follows:

[0058] 1) Calculate the "Hamming distance" through the hash algorithm to obtain the hash code sequence of each picture, each sequence has 64 characters;

[0059] 2) Compare the captured image I1 and the sample image I obtained by the same selection box m1 If the number of different characters in the two sequences is less than 5, the images are considered similar; otherwise, the images are considered dissimilar.

[0060] In the present invention, the position of the selection box in the standard template can be set by oneself. There are four selection boxes in the standard template, which are generally selected at the position with the indicative icon, that is, the position of the logo. The template examples of the four selection boxes in Table 1. The dotted line indicates the position of the logo in the boot process screen. The use of the selection box, firstly, takes into account the spatial distribution factors of the pixels in the image that will affect the histogram statistics; secondly, it greatly reduces the amount of calculation of the algorithm and shortens the time.

[0061] The starting position of the selection box is fixed, but it will be randomly mirrored and flipped according to the number of iterations, which can effectively reduce the impact of the spatial distribution of image features. At the same time, different sizes will result in different histogram statistics, which can improve the accuracy of the matching results.

[0062] The length and width of the selection box sH*sW can be adaptively adjusted according to the size of the image captured by the camera or can be set manually. In this example, the length and width of the image captured by the camera is H*W=300*300. Based on a random seed, any integer n between 3-5 is generated. The length and width of the selection box sH*sW is equal to the length and width of the image H*W divided by n, formula:

[0063] sH=H / n;sW=W / n

[0064] In this example, the length and width of selection box 1 are 100*100, that is, n=3 is randomly generated, and the length and width of 100 are obtained by 300 / 3; similarly, selection boxes 2, 3, and 4 are shown in Table 2.

[0065] Table 2 Calculation process of automatic adjustment of the length and width of the selection box

[0066]

[0067] The above is only a preferred implementation method of the method disclosed in the present invention to assist R&D personnel in analyzing the problem of a small probability of not starting up, and is not intended to limit the protection scope of the embodiments of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of this specification shall be included in the protection scope of the embodiments of this specification.

[0068] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0069] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0070] Each embodiment in the embodiments of this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

Claims

1. A method for assisting R&D personnel in analyzing a small probability of a computer not starting up, characterized in that: The steps include: The device under test is connected to the power supply and powered on; According to the preset power-on holding time, countdown and power-on and power-off test times, the power socket is controlled to be turned on or off, so that the device to be tested is in the power-on or power-off state; After the device under test is switched from a shutdown state to a startup state each time, a startup process video of the device under test is captured by an image capture device; Extract the boot process screen from the boot process video, the boot process screen includes the boot initial stage screen, the screen before entering the operating system, the screen for booting into the operating system and the screen for successfully entering the operating system, detect the boot process screen by digital image processing method, if it meets the detection specifications, save the detection result and continue to the next power on / off detection; if it does not meet the detection specifications, save the failed boot process video of the device to be tested, wherein the boot process screen is detected by digital image processing method whether it meets the specifications, and the specific steps are as follows: extract an image block with an indicative icon from the boot process screen; cut the image block according to a standard template to obtain m area images, the standard template is a selection box with different sizes; perform similarity comparison and matching on the m area images and the corresponding area images in the sample image in turn, if the comparison results of the m area images are all less than a preset threshold, perform similarity comparison and matching on the next image block, until the comparison results of all image blocks are less than the preset threshold, then it is determined that the boot process screen meets the specifications; Otherwise, it is determined that the boot process screen does not meet the detection specifications.

2. The method for assisting R&D personnel in analyzing a small probability of a computer not starting up according to claim 1, characterized in that: The method further includes the following steps: counting the power on / off state of the device under test, and when the device under test switches from the power off state to the power on state, the count is increased by 1; when the counted number is equal to the preset power on / off test number, the power on / off test is stopped.

3. The method for assisting R&D personnel in analyzing a small probability of a computer not starting up according to claim 1, characterized in that: The method also includes the following steps: detecting the boot process screen by a digital image processing method, and issuing a prompt alarm if the screen does not meet the detection specifications.

4. The method for assisting R&D personnel in analyzing a small probability of a computer not starting up according to claim 1, characterized in that: The similarity comparison and matching adopts a similarity matching method of hash coding and color histogram.

5. The method for assisting R&D personnel in analyzing a small probability of a computer not starting up according to claim 4, characterized in that: The boot process screen is pre-processed.

6. The method for assisting R&D personnel in analyzing a small probability of a computer not starting up according to claim 5, characterized in that: The preprocessing is median filtering.