TWI715193B - Remote invigilation system and method thereof - Google Patents

Abstract

A remote invigilation method includes the following steps. The subject host sends an operation data record and an operation video record to a data analysis host, and then the inspector host determines whether the operation data record and the operation video record contain an illegal signal. When the data analysis host determines that at least one of the operation data record and the operation video record includes the illegal signal, an alert signal is generated and sent to an invigilator host. The invention is more related to a remote invigilation system operating in the manner described above. The remote invigilation system and the method thereof provided by the invention can effectively reduce the labor cost required for the computer practical operation test.

Description

Remote invigilation system and method

The present invention relates to a remote invigilation system and method, in particular to a remote invigilation system and method suitable for computer practice tests.

Today, many practical courses have used Computer Practice Test (CPT) to replace the traditional paper-and-pencil test. For the test subjects, because they need to operate the designated computer software (such as Adobe XD or Photoshop) to complete the items required by the question during the test, they can verify their own performance more effectively than the paper and pen test. learning result. On the other hand, for lecturers, through the results of the computer practice test, they can also more effectively understand the proficiency of the testee's operational skills.

However, the difficulty of the computer practice test is to maintain the fairness of the test. For example, in order to prevent the testee from sending messages or answers through webpages, communication software, or mobile phones and other network tools, a common practice is to turn off the network during the test, but it also prevents the testee from real-time operation steps. Send to the invigilator. In addition, part of the online test method is to provide a stand-alone program for the tester. The stand-alone program will cover the entire screen of the tester. All tests must be completed in the stand-alone program; therefore, the above method is not applicable When you need to open more than one computer software operation test. On the other hand, some tests use a camera to record the test process, and this method still requires the invigilators to spend more time reviewing them one by one. Although some computer practical operation tests use manual monitoring and recording the operation screen of the test terminal to maintain fairness, it is difficult for invigilators to do both for classes with a large number of people.

The present invention provides a remote invigilation system and method. In addition to recording the image of the operation process, the system and method can also use image recognition technology to automatically determine whether the recorded image of the operation process violates regulations, so that the invigilator can When the testee violates the regulations, he immediately confirms and reacts to the violations, so that the invigilators can effectively grasp the situation of the examination room and improve the problems mentioned in the previous technology.

The remote invigilation method provided by the present invention includes the following steps. The host of the test-taking side transmits the operation data record and the operation image record to the data analysis end host, and then the data analysis end host determines whether the operation data record and the operation image record contain violation signals. Wherein, when the data analysis end host determines that at least one of the operation data record and the operation image record contains a violation signal, a warning signal is generated to the invigilation end host.

In an embodiment of the present invention, the above-mentioned data analysis terminal host determines whether the operation data record or the operation image record contains a violation signal, including the following steps. The data analysis end host obtains the execution program of the test-taking host according to the operation data record, and then the data analysis end host determines whether the execution program is in the first preset list. When the execution program is in the first preset list, the data analysis terminal host determines that the operation data record contains a violation signal. On the contrary, when the execution program is in the first preset list, the data analysis terminal host determines whether the operation data record contains URL information. When the operation data record contains URL information, the data analysis terminal host determines whether the URL information is in the second preset list. Conversely, when the web address information is not in the second preset list, the data analysis terminal host determines that the operation data record contains a violation signal.

In an embodiment of the present invention, the aforementioned test-taking host has a preset operation interface, and the operation image record optionally includes the preset operation interface, and the data analysis end host determines whether the operation data record or the operation image record contains a violation signal. Contains the following steps. The data analysis terminal host detects the default operation interface from the operation image record, and uses the operation image record containing the default operation interface as the image record to be analyzed. Following the previous step, the data analysis end host determines whether the operation data record is associated with the image record to be analyzed. When the operation data record is associated with the image record to be analyzed, The data analysis terminal host generates the test process record, where the test process record includes the image record to be analyzed and the operation data record. After the data analysis end host generates the test process records, the data analysis end host obtains these test process records within a preset period of time, and generates the test action rules accordingly. After the data analysis end host generates the tested action rule, the data analysis end host generates another tested process record and determines whether the similarity between the other tested process record and the tested action rule exceeds a preset value. When the similarity between another tested process record and the tested action rule does not exceed the preset value, the data analysis terminal host determines that the operation data record contains a violation signal.

In an embodiment of the present invention, the above-mentioned operation image record includes an image of the test subject, which is obtained by the recording component of the test taker host, and the data analysis end host determines whether the operation data record or the operation image record contains violations The signal includes the following steps. The host of the data analysis terminal detects the test object in the subject's image and determines whether the test object is in the third preset list. When the test object is in the third preset list, the data analysis terminal host determines that the operation image record contains a violation signal.

In an embodiment of the present invention, the following steps are included before the above-mentioned test-taking terminal host transmits the operation data record and the operation image record to the data analysis terminal host. The test taker host sends a login request to the data analysis end host, where the login request includes the address information of the test taker host, and the data analysis end host determines whether the address information is in the fourth preset list. When the address information is not in the fourth preset list, the data analysis host rejects the login request. Conversely, when the address information is in the fourth preset list, the data analysis host agrees to the login request.

The remote test invigilation system provided by the present invention includes a test-taking terminal host and a data analysis terminal host. The test-taking host includes an operating component, a data collection module, an image capture module, and a first network component, and the data analysis host includes a second network component and a judgment module. The operating component is configured to generate a trigger signal when triggered. The data collection module is configured to receive the trigger signal and generate operation data records accordingly, and the image capture module is configured to generate operation image records. The first network component is electrically connected to the data collection module and the image capture module to send out operation data records and operation image records. The data analysis terminal host includes a second network component and a judgment module. second The network component is connected to the first network component of the test terminal host, and is configured to receive operation data records and operation image records. The judgment module is electrically connected to the second network component and is configured to judge whether the operation data record and the operation image record contain a violation signal, wherein the judgment module is configured to determine whether at least one of the operation data record and the operation image record contains the violation signal, Generate a warning signal to the invigilator host.

In an embodiment of the present invention, the aforementioned data analysis terminal host further includes a database. The database stores a plurality of preset lists, and the plurality of preset lists are used by the judgment module to determine whether the operation data record and the operation image record are Contains violation signals.

In an embodiment of the present invention, the above-mentioned operation data record is associated with at least one of the executable program or web address information of the test taker host, and the data analysis end host is further configured to receive a login request from the test taker host. The login request includes the test taker The address information of the end host, and the judgment module includes an operation data record analysis unit configured to judge whether the execution program, the URL information, and the address information are in these preset lists.

In an embodiment of the present invention, the above-mentioned operation image record includes the default operation interface of the test-taking terminal host, the operation data record includes the use behavior of the preset operation interface, and the judgment module includes a use behavior analysis unit configured to analyze the preset The use behavior of the operating interface, and the use behavior is for the judgment module to determine whether the operating data record contains a violation signal.

In an embodiment of the present invention, the above-mentioned image capture module includes a recording component configured to record an image of a subject, and the judgment module includes an operation image analysis unit configured to detect and compare the image of the subject Whether the test object is in multiple preset lists.

Due to the adoption of the remote invigilation system and its method, the present invention can use image recognition technology to automatically determine whether the recorded image of the operation process violates regulations. During the test, the invigilator can immediately confirm and reflect the violations when the testees violate the rules. Therefore, through the system and method provided by the present invention, invigilators can more effectively grasp the condition of the examination room and maintain the fairness of the test.

In order to make the above and other objects, features and advantages of the present invention more obvious and understandable, the following Specific embodiments, in conjunction with the accompanying drawings, are described in detail as follows.

1: Test-taken host

11: Operating components

12: Data collection module

13: Image capture module

131: Video component

14: The first network component

2: Data analysis host

21: The second network component

22: Judgment module

221: Operation data record analysis unit

223: Use behavior analysis unit

222: Operating image analysis unit

23: Database

3: Invigilator host

31: screen

32: Warning component

S1, S2, S3, S21, S22, S23, S24, S25, S26, S27, S28, S29, S30, S31, S2a, S2b, S2c, S2d, S2e: steps

Fig. 1 is a block diagram of a remote invigilation system according to the first embodiment of the present invention.

Fig. 2 is a flowchart of the remote invigilation method according to the first embodiment of the present invention.

Fig. 3 is a block diagram of the remote invigilation system according to the second embodiment of the present invention.

Fig. 4 is a flowchart of the remote invigilation method according to the second embodiment of the present invention.

Fig. 5 is a block diagram of a remote invigilation system according to a third embodiment of the present invention.

Fig. 6 is a flowchart of a remote invigilation method according to the third embodiment of the present invention.

Fig. 7 is a flowchart of a remote invigilation method according to a fourth embodiment of the present invention.

Fig. 1 is a block diagram of a remote invigilation system according to the first embodiment of the present invention. Fig. 2 is a flowchart of the remote invigilation method according to the first embodiment of the present invention.

Please refer to Figure 1. The remote invigilation system includes the test-taking terminal host 1, the data analysis terminal host 2 and the invigilating terminal host 3. The test taker host 1 includes an operating component 11, a data collection module 12, an image capture module 13 and a first network component 14. The data analysis terminal host 2 includes a second network component 21 and a judgment module 22. The invigilator host 3 includes a screen 31 and a warning component 32. In detail, the test taker host 1 and the invigilator host 3 can be a personal computer, a notebook computer or a tablet; the data analysis host 2 can be a server or a personal computer, but the invention is not limited to this.

The operating component 11 is configured to generate a trigger signal when triggered. In practice, the operating component 11 can be a mouse and a keyboard for the subject to operate. On the other hand, when the test taker host 1 has a touch screen, the touch screen can also be used as the operating component 11 and generate a trigger signal when touched.

The image capturing module 13 can obtain a dynamic image of the test subject, such as a screen image or a video of the test subject, and generate an operation image record accordingly. For example, the image capture model Group 13 can be implemented with an application and an external camera, but it is not limited to this. In practice, the image information recorded by the image capture module 13 can be added to the corresponding time (time), and sent to the data analysis end host 2 via the network to facilitate subsequent analysis operations (please refer to Table 1) .

The data collection module 12 is electrically connected to the operating component 11, and is configured to receive a trigger signal and generate operating data records accordingly. The data collection module 12 can determine the action performed by the test subject on the operating component 11 through the trigger signal, and present the above-mentioned execution action with the operating data record. For example, the operation data record may include actions such as clicking, double-clicking, long-pressing, dragging or moving the cursor of the left or right mouse button, and may also include clicking actions of various character keys or function keys on the keyboard. Please refer to Table 2 for some examples, but this embodiment is not limited thereto.

It is worth mentioning that, in order to enable the test taker host 1 to continuously obtain operating data records, the test taker host 1 can continuously detect whether the data collection module 12 is operating normally, and issue a warning signal when the data collection module 12 stops operating. . For example, since the test terminal host 1 needs to install a test system for the tester to perform the test, the test system can be set to continuously exchange messages with the data collection module 12 through the HyperText Transfer Protocol (HTTP). To detect the operation of the data collection module 12. In addition, the warning signal can also be set to stop the test system (for example, turn off the question display screen) to remind the testee to turn on the data collection module 12.

The first network component 14 is electrically connected to the data collection module 12 and the image capture module 13 to send out operation data records and operation image records. The first network component 14 can be implemented as a wireless communication component, for example, a wireless network communication component. In addition, the first network component 14 can also be implemented as a wired communication component, for example, a network component with a network hole.

The second network component 21 of the data analysis host 2 is communicatively connected to the first network component 14 of the test-taking host 1, and is configured to receive operation data records and operation image records. Specifically, the second network component 2 has the same communication protocol as the first network component 14, so that the test taker host 1 can transmit the operation data record and the operation image record to the data analysis end host through the communication protocol. 2. Since the implementation of the second network component 2 is the same as that of the first network component 14, it will not be repeated here.

The judgment module 22 of the data analysis terminal host 2 is electrically connected to the second network component 21. The judgment module 22 is configured to judge whether the operation data record and the operation image record contain a violation signal, wherein the judgment module 22 generates a warning signal when one of the operation data record and the operation image record contains a violation signal. Specifically, the judgment module 22 can be an application program or software.

The invigilator host 3 can communicate with the data analysis host 2 via a network component (not shown) to receive the operation data records, operation image records, and warning signals sent by the data analysis host 2. The test invigilator host 3 is for invigilators to inspect the conditions of the examination room, so the test invigilator host 3 at least has a screen 31 and a warning component 32. In detail, the screen 31 and the warning component 32 are electrically connected to each other. The screen 31 is configured to present operation data records and operation image records, and the warning component 32 is configured to present warning signals. In practice, the warning signal can be presented by sound, video, text message or flashing lights. Therefore, according to different types of warning signals, the warning component 32 can be configured as a speaker or a lamp, and can also be configured as a signal transceiver electrically connected to the screen 31.

Please refer to Figure 2. First, refer to step S1: the test-taking host 1 transmits the operation data record and the operation image record to the data analysis end host 2; the operation data record and the type and time that can be operated are presented in a list, and the operation image record can be according to the testee Information such as account number and question number are presented in a list. For the continuation, please refer to step S2: The data analysis terminal host 2 determines whether the operation data record and the operation image record contain violation signals. For example, the violation signal can be related to whether the test-taking host 1 has opened an application or website outside the standard, whether the test-taking host 1 has not operated the application specified by the test for a long time, or whether the testee uses an object outside the standard to take the test, etc. However, this embodiment is not limited to this.

When the data analysis terminal host 2 determines that at least one of the operation data record and the operation image record contains a violation signal, please refer to step S3: the data analysis terminal host 2 generates a warning signal; wherein the warning signal is sent to the invigilation terminal host 3. In practice, the warning signal can be set to simultaneously interrupt the running test of the data analysis terminal host 2 that generates the violation signal to respond to the violation in real time. Conversely, when the data analysis terminal host 2 determines that neither the operation data record nor the operation image record contains a violation signal, it returns to step S1.

Fig. 3 is a block diagram of the remote invigilation system according to the second embodiment of the present invention. Fig. 4 is a flowchart of the remote invigilation method according to the second embodiment of the present invention.

Please refer to Figure 3. In this embodiment, the aforementioned data analysis terminal host 2 further includes a database 23. The database 23 stores a plurality of preset lists, and these preset lists are used by the judgment module 22 to judge whether the operation data record and the operation image record contain violation signals. In detail, these default lists are related to multiple test criteria, such as applications that are not allowed in the test, websites that allow links, applications that are allowed to complete test questions, and test objects that are allowed to be used in tests Wait.

Please continue to refer to Figure 3. In this embodiment, the above-mentioned operation data record is associated with at least one of the executable program or the website information of the test taker host 1. The data analysis end host 2 is further configured to receive a login request from the test taker host 1, and the login request includes the address information of the test taker host 1. The judgment module 22 includes an operation data record analysis unit 221 configured to judge execution Whether the program, URL information, and address information are in these default lists. In practice, the operation data record analysis unit 221 may be an application program or program code of the data analysis terminal host 2, and can determine the application program and web page used by the test-taking terminal host from the operation data record, as described below for details.

Please refer to steps S21 to S25 in FIG. 4. Please refer to step S21: the data analysis end host 2 obtains the execution program of the test candidate host 1 according to the operation data record; the execution program may be an application that has been opened by the data analysis end host 2. Please proceed to step S22: The data analysis terminal host 2 determines whether the execution program is in the first preset list, where the first preset list contains applications that cannot be used in the test, such as communication software such as Line, Message, and Skype. When the execution program is in the first preset list, please refer to step S23: The data analysis terminal host 2 determines that the operation data record contains a violation signal, and can further generate a warning signal to remind the invigilator of the violation action in the examination room in real time. Conversely, when the running program is not in the first preset list, please refer to step S24: The data analysis terminal host 2 determines whether the operation data record contains URL information; the URL information may be the information of the webpage opened by the data analysis host 2 URL. When the operation data record contains web address information, please refer to step S25: The data analysis terminal host 2 determines whether the web address information is in the second preset list, where the second preset list includes websites that are allowed to be visited in the test. When the data analysis terminal host 2 determines that the web address information is not in the second preset list, the above step S23 is continued. Otherwise, go back to step S21.

Please refer to steps S23, S26 and S27 in FIG. 4. First, refer to step S26: the data analysis terminal host 2 detects the test object in the testee's image, where the test object can be a mobile phone, an external hard disk, a portable hard disk, or a reference book. Please proceed to step S27: The data analysis terminal host 2 determines whether the test object is in the third preset list, and the third preset list is associated with the test object that is not allowed to be used in the test. For example, if the use of mobile phones is not allowed during the test, the third preset list can pre-store various types of mobile phone images for the data analysis terminal host 2 to determine from the testee’s images whether the testee is using a mobile phone to take the test . Similarly, according to different test rules, the test object may also be written materials such as notes, lecture notes or reference books, but the present invention is not limited thereto. When the test object is in the third preset list, please refer to step S23. Because step S23 has been explained in the previous section, Therefore, I will not repeat them here. Otherwise, return to step S26.

Fig. 5 is a block diagram of a remote invigilation system according to a third embodiment of the present invention. Fig. 6 is a flowchart of a remote invigilation method according to the third embodiment of the present invention.

Please refer to Figure 5. In one embodiment, the aforementioned image capturing module 13 includes a recording component 131 configured to record images of the subject. The image of the subject is used by the judging module 22 to detect and compare whether the corresponding test object is in the preset list. Specifically, the recording component 131 can be any camera that can be electrically connected to the first network component 14, for example, a webcam built in or external to the test taker host 1 to record the test during the test. The image of the subject.

In this embodiment, the judgment module 22 of the data analysis terminal host 2 includes an operation image analysis unit 222 and a usage behavior analysis unit 223.

The operation image analysis unit 222 is configured to detect and compare whether the test object in the subject image is in a plurality of preset lists. For example, the operation image analysis unit 222 may be an application program installed in the test taker host 1 and has the function of recording screen moving images. Please see below for detailed operation.

In this embodiment, the operation image record includes the default operation interface of the test-taking terminal host 1, and the operation data record includes the usage behavior of the default operation interface. As mentioned above, the usage behavior analysis unit 223 is configured to analyze the usage behavior of the preset operation interface, and the usage behavior is used by the judgment module 22 to determine whether the operation data record contains a violation signal. In detail, the test-taking terminal host 1 has a preset operation interface, and the operation image record optionally includes the preset operation interface. For example, when the test taker host 1 is using the default operation interface for drawing or writing programs, the operation image record will include the default operation interface; on the contrary, when the test taker host 1 is using other programs (such as using a web browser) When querying data), because the window of the default operation interface may be temporarily minimized, the operation image record does not include the default operation interface. The data analysis terminal host 2 can judge from the above-mentioned operation image records whether the test-taking host 1 actually uses the designated software to generate test results during the test. In practice, the default operation interface can be an application program for specified operations; and the usage behavior analysis unit 223 can be software or hardware with computing functions, but the present invention does not This is limited. Please see below for the detailed operation of the above components.

Please refer to steps S2a to S2e and S23 in FIG. 6. First, refer to step S2a: the data analysis terminal host 2 detects the preset operation interface from the operation image record, and uses the operation image record containing the preset operation interface as the image record to be analyzed. To put it simply, this step is to first filter the operation image records that do not contain the default operation interface, so as to effectively use the computing performance of the data analysis terminal host 2. Please refer to step S2b: The data analysis terminal host 2 determines whether the operation data record is associated with the image record to be analyzed. For example, the operation data record may include the time when the operation action of the test-taking host 1 occurs, and the coordinates corresponding to the action on the screen (part of the operation action can be referred to Table 2 above). Therefore, the data analysis terminal host 2 can further determine whether the operation action is performed in the default operation interface by comparing the above information.

As mentioned above, when the operation data record is associated with the image record to be analyzed, please refer to step S2c: The data analysis terminal host 2 generates the test process record; the test process record includes the image record to be analyzed and the operation data record. For example, the test process record can be presented as T={(u1;a11,...,a1m),...,(un;an1,...,anm)}, where u1 is the default operation interface corresponding to different test questions, a11 To a1m is the operation action generated by host 1 of the test-taking terminal to u1. After generating the tested process records, please proceed to step S2d: The data analysis terminal host 2 obtains multiple tested process records within a preset time period, and generates the tested action rules based on them; the tested action rules can be regarded as most tested The overall operation behavior of the user on the preset operation interface. For example, the data analysis terminal host 2 can integrate all the collected test process records into test action rules within half an hour of the start of the test, so as to serve as a benchmark for all testees' operational behavior. It is worth mentioning that the data analysis end host 2 can use Association Rule Mining, according to the parameters specified by the system, including minimum support (minimum support) and minimum confidence (minimum confidence), from These test process records find out the test action rules. In practice, the tested action rule can be presented as R={r1,...,rn}, but it is not limited to this.

After the data analysis end host 2 generates the tested action rule, please refer to step S2e: The data analysis end host 2 generates another test process record, and judges the other test process record and the test action Whether the similarity of the rules exceeds the preset value. Here, another test process record may be the latest test process record obtained by the data analysis terminal host 2 after the aforementioned preset time period. When the similarity between another tested process record and the tested action rule does not exceed the preset value, it means that the operation behavior of the test-taking host 1 is not a normal behavior. For example, within half an hour of the start of the test, the overall operation behavior of the tested action rule is "press and hold the left mouse button and drag the cursor", but the operation behavior presented by the test-taking host 1 is "intermittently single Click the left mouse button". At this time, because the data analysis host 2 determines that the operation behavior of the test-taking host 1 is too different from the other test-taking hosts 1, and there is a risk of violation, please proceed to step S23. Since step S23 has been described in the foregoing, it will not be repeated here. Otherwise, go back to step S2d.

Please refer to FIG. 7, which is a flowchart of a remote invigilation method according to a fourth embodiment of the present invention. In this embodiment, the test-taking host 1 must first log in to the test system of the data analysis host 2 before performing the test. Therefore, the data analysis host 2 can use the address information provided by the test-taking host 1 to log in to the test system to determine whether the testee uses the designated host to perform the test. Please refer to step S28: The test-taking host 1 sends a login request to the data analysis host 2. The login request includes the address information of the test taker host 1, and the address information can be the Internet Protocol Address (IP Address) or the default host number of the test taker host 1, but not Is limited. Please refer to step S29: The data analysis end host 2 determines whether the address information is in the fourth preset list, where the fourth preset list is associated with the test-taken host 1 in the specification. When the address information is not in the fourth preset list, please refer to step S30: The data analysis end host 2 rejects the login request; the data analysis end host 2 can send a warning message to the invigilator host 3 at the same time to notify the invigilator immediately. Participants use non-standard computers to take the test. Otherwise, refer to step S31: the data analysis end host 2 agrees to the login request, so that the test-taking host 1 can start the test (that is, step S1 in FIG. 2 is continued).

In summary, due to the use of the remote invigilation system and its method, the present invention can use image recognition technology to automatically determine whether the recorded image of the operation process violates regulations. During the test, the invigilator can immediately confirm and reflect the violations when the testees violate the rules. Therefore, through the system and method provided by the present invention, invigilators can more effectively grasp the condition of the examination room and maintain the fairness of the test.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to those defined by the attached patent application scope.

S1, S2, S3: steps

Claims (9)

A remote invigilation method includes the following steps: a test-taking host transmits an operation data record and an operation image record to a data analysis end host; and the data analysis end host determines whether the operation data record and the operation image record are Contains a violation signal; wherein, when the data analysis end host determines that at least one of the operation data record and the operation image record contains the violation signal, it generates a warning signal to an invigilation end host; wherein, the test-taking end host Having a preset operation interface, and the operation image record optionally includes the default operation interface, the data analysis terminal host determining whether the operation data record or the operation image record contains the violation signal includes the following steps: The data analysis end host detects the default operation interface from the operation image record, and uses the operation image record containing the default operation interface as an image record to be analyzed; the data analysis end host determines whether the operation data record is Associated with the image record to be analyzed; when the operation data record is associated with the image record to be analyzed, a measured process record is generated by the data analysis terminal host, wherein the measured process record includes the image record to be analyzed and the operation data Record; use the data analysis end host to obtain the test process records within a preset period of time, and generate a test action rule based on it; use the data analysis end host to generate another test process record and determine the other Whether the similarity between a tested process record and the tested action rule exceeds a preset value; and When the similarity between the other tested process record and the tested action rule does not exceed the preset value, the data analysis terminal host determines that the operation data record contains the violation signal. For example, the remote invigilation method described in item 1 of the request, wherein the data analysis end host determines whether the operation data record or the operation image record contains the violation signal, including the following steps: the data analysis end host is used according to the operation Data record, obtain an execution program of the test-taking host; use the data analysis host to determine whether the execution program is in a first preset list; when the execution program is in the first preset list, use the data to analyze The end host determines that the operation data record contains the violation signal; when the executable program is not in the first preset list, the data analysis end host determines whether the operation data record contains a URL information; when the operation data record contains the URL Information, the data analysis end host determines whether the web address information is in a second preset list; and when the web address information is not in the second preset list, the data analysis end host determines that the operation data record contains the Violation signal. Such as the remote invigilation method described in item 1 of the request, wherein the operation image record includes an image of the testee, the image of the testee is obtained by a recording component of the test-taking host, and the data analysis host determines Whether the operation data record or the operation image record contains the violation signal includes the following steps: use the data analysis terminal host to detect a test object in the subject’s image, and determine whether the test object is a third predictor Set in the list; and When the test object is in the third preset list, the data analysis end host determines that the operation image record contains the violation signal. For example, the remote invigilation method described in claim 1, wherein before the test taker host transmits the operation data record and the operation image record to the data analysis end host, it includes the following steps: the test taker host transmits a A login request to the data analysis end host, where the login request contains one address information of the test taker host; the data analysis end host is used to determine whether the address information is in a fourth preset list; when the address information If it is not in the fourth preset list, the data analysis end host rejects the login request; and when the address information is in the fourth preset list, the data analysis end host agrees to the login request. A remote invigilation system includes: an exam-taking host, including: an operating component configured to generate a trigger signal when triggered; a data collection module configured to receive the trigger signal and generate an operating data record accordingly ; And an image capture module configured to generate an operation image record; and a first network component electrically connected to the data collection module and the image capture module to send out the operation data record and the Operation image record; a data analysis terminal host, including: A second network component, communicatively connected to the first network component, and configured to receive the operation data record and the operation image record; a judgment module, electrically connected to the second network component, and configured to judge the Whether the operation data record and the operation image record contain a violation signal, wherein the judgment module generates a warning signal to an invigilator host when at least one of the operation data record and the operation image record contains the violation signal; Wherein, the test-taken host has a preset operating interface, and the operating image record optionally includes the preset operating interface. The data analysis host uses the following steps to determine whether the operating data record or the operating image record contains the Violation signal: use the data analysis end host to detect the default operation interface from the operation image record, and use the operation image record containing the default operation interface as an image record to be analyzed; judge by the data analysis end host Whether the operation data record is associated with the image record to be analyzed; when the operation data record is associated with the image record to be analyzed, the data analysis terminal host generates a measured process record, wherein the measured process record includes the image to be analyzed Records and the operation data records; use the data analysis terminal host to obtain the tested process records within a preset period of time, and generate a tested action rule accordingly; use the data analysis terminal host to generate another tested process record , And determine whether the similarity between the other tested process record and the tested action rule exceeds a preset value; and when the similarity between the other tested process record and the tested action rule does not exceed the preset value , The data analysis terminal host determines that the operation data record contains the violation signal. For example, the remote invigilation system according to claim 5, wherein the data analysis terminal host further includes a database, the database stores a plurality of preset lists, and the preset lists are used for the judgment module to judge Whether the operation data record and the operation image record contain the violation signal. For example, the remote test invigilation system according to claim 6, wherein the operation data record is associated with at least one of an executable program or a website information of the test taker host, and the data analysis end host is further configured to receive the test taker A login request from the host, the login request includes the address information of the test-taken host, and the determination module includes an operation data record analysis unit configured to determine whether the executable program, the URL information, and the address information are In these preset lists. The remote invigilation system according to claim 6, wherein the operation image record includes a preset operation interface of the test taker host, the operation data record includes a usage behavior of the preset operation interface, and the judgment module It includes a usage behavior analysis unit configured to analyze the usage behavior of the preset operation interface, and the usage behavior is used by the judgment module to judge whether the operation data record contains the violation signal. The remote invigilation system according to claim 6, wherein the image capture module includes a recording component configured to record an image of a subject, and the judgment module includes an operating image analysis unit configured to detect And compare whether a test object in the subject's image is in the preset lists.

Images