CN111241945A - Method and device for testing face recognition performance, computer equipment and storage medium - Google Patents
Method and device for testing face recognition performance, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111241945A CN111241945A CN201911418325.0A CN201911418325A CN111241945A CN 111241945 A CN111241945 A CN 111241945A CN 201911418325 A CN201911418325 A CN 201911418325A CN 111241945 A CN111241945 A CN 111241945A
- Authority
- CN
- China
- Prior art keywords
- face
- detection
- face detection
- static
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 116
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000001514 detection method Methods 0.000 claims abstract description 424
- 230000003068 static effect Effects 0.000 claims abstract description 120
- 238000001574 biopsy Methods 0.000 claims description 26
- 238000004458 analytical method Methods 0.000 claims description 24
- 238000001727 in vivo Methods 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 15
- 238000005286 illumination Methods 0.000 claims description 7
- 230000008921 facial expression Effects 0.000 claims description 5
- 238000003556 assay Methods 0.000 claims description 4
- 238000011056 performance test Methods 0.000 abstract description 20
- 238000010586 diagram Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 210000001508 eye Anatomy 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/40—Spoof detection, e.g. liveness detection
- G06V40/45—Detection of the body part being alive
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a method and a device for testing face recognition performance, computer equipment and a storage medium, wherein face detection is carried out on an acquired real-time streaming image according to a face detection algorithm; under the condition that the human face detection is judged not to pass according to the human face detection threshold value, the real-time streaming image is stored as a static human face detection image; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the same face detection rate according to the result of the static face detection; stopping the face detection under the condition that the total number of face detection times is greater than or equal to the preset number of face detection times, and outputting the same face detection rate and face detection data to a test log; and acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance, so that the problem of low test efficiency in the face recognition performance test is solved.
Description
Technical Field
The present application relates to the field of face recognition technology, and in particular, to a method and an apparatus for testing face recognition performance, a computer device, and a storage medium.
Background
With the continuous development of face recognition technology, various face recognition devices are widely used. Face identification equipment mainly used identification, for accurate identification, need carry out performance test to face identification equipment, for example, carry out the discernment precision test of each angle to the face identification equipment who develops, face identification equipment side after the test is qualified can put into production and make. In the related art, a method for testing face recognition equipment is to test the recognition performance of the equipment with a face recognition system by holding the equipment by a tester for face recognition verification, and the testing efficiency is low due to manual testing.
Aiming at the problem of low testing efficiency in a face recognition performance test in the related art, no effective solution is provided at present.
Disclosure of Invention
The invention provides a method and a device for testing face recognition performance, computer equipment and a storage medium, aiming at the problem of low testing efficiency in face recognition performance testing in the related art, and at least solving the problem.
According to an aspect of the present invention, there is provided a method for testing face recognition performance, the method including:
carrying out face detection on the acquired real-time stream image according to a face detection algorithm;
under the condition that the human face detection is judged not to pass according to a human face detection threshold value, the real-time streaming image is stored as a static human face detection image; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the face detection identity rate according to the result of the static face detection;
performing the face detection for multiple times, and outputting the face detection same rate and face detection data to a test log under the condition that the total number of the face detection is greater than or equal to the preset number of the face detection;
and acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance.
In one embodiment, after the performing face detection on the acquired real-time streaming image according to a face detection algorithm, the method further includes:
performing living body detection on the real-time stream image according to a living body detection algorithm under the condition that the face detection is passed;
saving the real-time stream image as a static biopsy image under the condition that the biopsy is judged not to pass according to a biopsy threshold; performing static living body detection on the static living body detection image according to the living body detection algorithm, and acquiring the same rate of the living body detection according to the result of the static living body detection;
performing the live body detection a plurality of times, and outputting the same rate of live body detection and live body detection data to the test log in the case that the total number of the live body detections is greater than or equal to the preset number of live body detections;
and acquiring a living body detection counting index according to the static living body detection image and the test log, and representing the face identification performance according to the living body detection counting index and the face detection counting index.
In one embodiment, after the live-detecting the live-streaming image according to the live-detecting algorithm, the method further comprises:
under the condition that the living body detection is passed, carrying out face comparison on the real-time stream image according to a face comparison algorithm;
under the condition that the face comparison is not passed according to a face comparison threshold value, storing the real-time streaming image as a static face comparison image; performing static face comparison on the static face comparison image according to the face comparison algorithm, and acquiring the face comparison identity rate according to the static face comparison result;
performing face comparison for multiple times, and outputting the same face comparison rate and face comparison data to the test log under the condition that the total number of face comparison is greater than or equal to the preset number of face comparison;
and acquiring a face comparison counting index according to the static face comparison image and the test log, and representing the face recognition performance according to the face comparison counting index, the living body detection counting index and the face detection counting index.
In one embodiment, the obtaining a liveness detection count indicator from the static liveness detection image and the test log comprises:
performing in-vivo detection analysis on the static in-vivo detection image, and acquiring the in-vivo detection counting index according to the in-vivo detection analysis and the test log; wherein the in vivo detection assay comprises: illumination analysis, paper analysis, and prosthesis analysis.
In one embodiment, the obtaining a detection count index according to the static face detection image and the test log includes:
performing image quality detection on the static face detection image, analyzing the test log according to the image quality detection, and acquiring the face detection counting index; wherein the image quality detection comprises at least one of: resolution, face integrity, face interpupillary distance, image sharpness, face expression, face angle, and image contrast.
In one embodiment, after obtaining the face detection count index according to the static face detection image and the test log, the method further includes:
sending the face detection counting index to a terminal; and the terminal displays the face detection counting index in a form of a chart.
According to another aspect of the present invention, there is provided an apparatus for testing face recognition performance, the apparatus comprising:
the face detection module is used for carrying out face detection on the acquired real-time stream image according to a face detection algorithm;
the storage module is used for storing the real-time streaming image as a static face detection image under the condition that the face detection is judged not to pass according to a face detection threshold value; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the face detection identity rate according to the result of the static face detection;
the output module is used for outputting the face detection identical rate and face detection data to a test log when the face detection is carried out for a plurality of times and the total number of times of the face detection is greater than or equal to the preset number of times of the face detection;
and the acquisition module is used for acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance.
In one embodiment, the apparatus further comprises a liveness detection module;
the living body detection module performs living body detection on the real-time stream image according to a living body detection algorithm under the condition that the face detection is passed;
the storage module stores the real-time stream image as a static living body detection image under the condition that the living body detection is judged not to pass according to a living body detection threshold; performing static living body detection on the static living body detection image according to the living body detection algorithm, and acquiring the same rate of the living body detection according to the result of the static living body detection;
the output module performs the living body detection for a plurality of times, and outputs the same living body detection rate and living body detection data to the test log when the total number of the living body detections is greater than or equal to the preset number of living body detections;
and the performance module acquires a living body detection counting index according to the static living body detection image and the test log, and represents the face recognition performance according to the living body detection counting index and the face detection counting index.
According to another aspect of the present invention, there is provided a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the methods described above when executing the computer program.
According to another aspect of the invention, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of any of the methods described above.
According to the invention, the face detection is carried out on the acquired real-time streaming image according to the face detection algorithm by adopting the method and the device for testing the face recognition performance, the computer equipment and the storage medium; under the condition that the human face detection is judged not to pass according to the human face detection threshold value, the real-time streaming image is stored as a static human face detection image; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the same face detection rate according to the result of the static face detection; stopping the face detection under the condition that the total number of the face detection times is greater than or equal to the preset number of the face detection times, and outputting the same face detection rate and face detection data to a test log; and acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance, so that the problem of low test efficiency in the face recognition performance test is solved.
Drawings
FIG. 1 is a schematic diagram of an application scenario of a face recognition performance test according to an embodiment of the present invention;
FIG. 2 is a first flowchart of a method for testing face recognition performance according to an embodiment of the present invention;
FIG. 3 is a flowchart II of a face recognition performance testing method according to an embodiment of the present invention;
fig. 4 is a flowchart three of a face recognition performance testing method according to an embodiment of the present invention;
FIG. 5 is a fourth flowchart of a face recognition performance testing method according to an embodiment of the present invention;
FIG. 6 is a fifth flowchart of a method for testing face recognition performance according to an embodiment of the present invention;
FIG. 7 is a sixth flowchart of a method for testing face recognition performance according to an embodiment of the present invention;
FIG. 8 is a block diagram of a first exemplary embodiment of a face recognition performance testing apparatus;
FIG. 9 is a block diagram of a second embodiment of a face recognition performance testing apparatus according to the present invention;
fig. 10 is a block diagram of a structure of a face recognition performance testing apparatus according to an embodiment of the present invention;
fig. 11 is a block diagram of the inside of a computer device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In this embodiment, a method for comparing human faces is provided, and fig. 1 is a schematic diagram of an application scenario of a human face recognition performance test according to an embodiment of the present invention, as shown in fig. 1, in the application environment, a terminal 12 communicates with a server 14 through a network. The server 14 performs face detection on the acquired real-time streaming image according to a face detection algorithm; the server 14 stores the real-time streaming image as a static face detection image and obtains the same face detection rate under the condition that the face detection is judged not to pass according to the face detection threshold; the server 14 outputs the face detection identity rate and the face detection data to a test log, and obtains a face detection counting index according to the static face detection image and the test log; the terminal 12 receives and displays the face detection count index sent by the server 14. The terminal 12 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 14 may be implemented by an independent server or a server cluster composed of a plurality of servers.
In this embodiment, a method for testing a face recognition performance is provided, and fig. 2 is a first flowchart of a method for testing a face recognition performance according to an embodiment of the present invention, as shown in fig. 2, the method includes the following steps:
step S202, acquiring dynamic real-time stream images, and carrying out face detection on each acquired real-time stream image by calling an interface of a face detection algorithm; the face detection algorithm can be realized by methods such as two-wavelet transformation, an elastic model or a neural network.
Step S204, under the condition that the human face detection is judged not to pass according to the human face detection threshold value, the real-time stream image is stored as a static human face detection image; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the same face detection rate according to the result of the static face detection; the face detection threshold value can be preset by a tester according to actual test requirements.
The method comprises the steps that real-time streaming images of undetected faces are stored as static face detection images, and the static face detection images can be divided into infrared images, color images and depth images and are stored in different files; and calling an algorithm interface again for the static face detection image, judging whether a static detection result is consistent with a result obtained in dynamic detection, and if not, checking or correcting the face detection algorithm.
Step S206, the face detection is carried out for a plurality of times, under the condition that the total number of the face detection is more than or equal to the preset number of the face detection, namely when the preset number of the face detection appointed by a user is reached, the real-time streaming image is stopped to be stored, the face detection identical rate and the face detection data are output to a test log, and the checking is more intuitive and convenient; the face detection identity rate is used for representing the accuracy of the face detection performance; the face detection data includes: total detection times, the times of detecting the human face and the times of not detecting the human face.
Step S208, acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance; the face detection count index may be obtained by direct statistics from the test log, and the face detection count index includes: detecting a stream frame rate, a face detection passing rate, a face detection rejection rate and a face false detection rate; detecting the flow frame rate as the total frame number/testing duration of the face detection; the face detection passing rate is the number of times of detecting faces/total number of frames of face detection; the face detection rejection rate is the number of times that the face is not detected/the total number of frames of face detection; the face false detection rate is the face false detection times/the face detection times. The higher the face detection passing rate is, and the lower the face detection rejection rate and the face false detection rate are, the better the face detection performance is; the lower the face detection passing rate is, and the higher the face detection rejection rate and the face false detection rate are, the lower the face detection performance is.
In the related art, the performance of the face recognition device needs to be tested manually, but in the embodiment of the invention, through the steps S202 to S208, the face detection is performed on the acquired real-time stream image according to the face detection algorithm, the real-time stream image which fails in the face detection is stored as a static image, and the static image is detected and confirmed again, so that the reason why the face is not detected is conveniently and quickly analyzed; meanwhile, a face detection counting index is obtained according to the face detection data and is used for representing the face recognition performance, so that the face recognition performance is automatically tested, and the problem of low testing efficiency in the face recognition performance test is solved.
In an embodiment, a method for testing a face recognition performance is provided, and fig. 3 is a flowchart of a method for testing a face recognition performance according to an embodiment of the present invention, as shown in fig. 3, the method includes the following steps:
step S302, carrying out living body detection on the real-time stream image according to a living body detection algorithm under the condition that the face detection is passed; whether the detected object is an animate individual can be determined through the living body detection, and the detected object is not an inanimate object such as a photo, a video and the like, so that a malicious attacker can be prevented from carrying out malicious attack in the modes of a recorded video, a shot photo, a 3D human face model, a forged mask and the like.
Step S304, under the condition that the living body detection threshold value is used for judging that the living body detection is not passed, the real-time stream image is saved as a static living body detection image; wherein, the static living body detection image can be divided into an infrared image, a color image and a depth image and stored in different files.
Performing static biopsy on the static biopsy image according to the biopsy algorithm, and acquiring the biopsy identity rate according to the result of the static biopsy; wherein, the judgment of living body and non-living body is carried out by utilizing the infrared image, the color image and the depth image; whether the living body is erroneously determined to be a non-living body due to the influence of the light is checked based on the color image.
Step S306, the live body detection is carried out for a plurality of times, the real-time stream image is stopped to be stored under the condition that the total times of the live body detection is greater than or equal to the preset times of the live body detection, and the same rate of the live body detection and the data of the live body detection are output to the test log; wherein the in-vivo detection data includes: total number of detections, number of detections of living organisms, and number of non-detections of living organisms.
Step S308, acquiring a living body detection counting index according to the static living body detection image and the test log, and representing the face recognition performance according to the living body detection counting index and the face detection counting index; wherein, the living body detection counting index comprises: a living error acceptance rate and a living error rejection rate; the living body false acceptance rate is the percentage of the number of the face living bodies which are judged by mistake in the living body detection process to the number of the false bodies which are to be identified in the test set, the living body false acceptance rate and the living body false rejection rate simultaneously meet the conditions that FAR is less than 0.1 percent and FRR is less than or equal to 1 percent; the in-vivo false rejection rate is the percentage of the number of false human face prostheses judged in the in-vivo detection process to the number of tests to be identified as the in-vivo in the test set.
Through the steps S302 to S308, the passing rate and the rejection rate of face recognition are respectively tested and recorded in different scenes, under the condition that the living body detection is judged not to pass according to the living body detection threshold, the infrared images, the color images and the depth images of the undetected faces and the images of the non-living bodies are respectively and simultaneously stored, aiming at the images of the undetected faces and the images of the non-living bodies, secondary detection and confirmation are carried out by calling a face detection algorithm interface, and further analysis and classification are carried out to investigate reasons, so that a tester can carry out rapid analysis conveniently; meanwhile, a living body detection counting index is obtained according to the living body detection data, and the face recognition performance is represented according to the living body detection counting index and the face detection counting index, so that the efficiency and the comprehensiveness of the face recognition performance test are improved.
In an embodiment, a method for testing a face recognition performance is provided, and fig. 4 is a flowchart three of the method for testing a face recognition performance according to the embodiment of the present invention, as shown in fig. 4, the method includes the following steps:
step S402, extracting 10000 pieces of face photo test data of different orientation, distance, expression, shielding and illumination environments of a local picture under the condition that the living body detection passes, including target pictures (including two same targets), performing face detection and feature extraction, and registering the face photo test data in a feature database; comparing the face of the real-time streaming image according to a face comparison algorithm and the feature database; wherein, the mode of the face comparison comprises: color image face comparison, infrared image face comparison, 1:1 comparison mode, 1: N comparison mode or M: N mode; the face comparison algorithm can be realized through a neural network.
Step S404, under the condition that the face comparison is judged not to pass according to the face comparison threshold value, the real-time streaming image is saved as a static face comparison image; and carrying out static face comparison on the static face comparison image according to the face comparison algorithm, and obtaining the face comparison identical rate according to the static face comparison result.
Step S406, comparing the human face for a plurality of times, stopping the human face comparison under the condition that the total times of the human face comparison is greater than or equal to the preset human face comparison times, and outputting the human face comparison identical rate and human face comparison data to the test log; wherein, the face comparison data comprises: the total number of comparison, the number of comparison passing and the number of comparison false alarm.
Step S408, acquiring a face comparison counting index according to the static face comparison image and the test log, and representing the face recognition performance according to the face comparison counting index, the living body detection counting index and the face detection counting index; wherein, the face comparison counting index comprises: comparing the passing rate with the false alarm rate; comparing the passing rate to the total number of times of comparison; and comparing the false alarm rate, namely comparing the false alarm times/the comparison passing times. The face comparison performance test table is shown in table 1:
table 1 face comparison performance test table
Through the steps S402 to S408, the real-time streaming image is saved as a static face comparison image under the condition that the face comparison is not passed according to the face comparison threshold, so that the comparison effect can be analyzed quickly; and acquiring a face comparison counting index, representing the face recognition performance according to the face comparison counting index, the living body detection counting index and the face detection counting index, expanding the scene of face recognition performance test, and further improving the efficiency, accuracy and comprehensiveness of the face recognition performance test.
In an embodiment, a method for testing a face recognition performance is provided, and fig. 5 is a fourth flowchart of the method for testing a face recognition performance according to the embodiment of the present invention, as shown in fig. 5, the method includes the following steps:
step S502, performing in-vivo detection analysis on the static in-vivo detection image, and acquiring the in-vivo detection counting index according to the in-vivo detection analysis and the test log; wherein the in vivo detection assay comprises: illumination analysis, paper analysis and prosthesis analysis; wherein, the in vivo examination performance test table is shown in table 2:
table 2 in vivo examination performance test table
Through the step S502, the static living body detection image is subjected to living body detection analysis, and a living body detection counting index is obtained, so that the problem positioning and the rapid analysis of the activity detection effect are realized, and the efficiency of the face recognition performance test is improved.
In an embodiment, a method for testing a face recognition performance is provided, and fig. 6 is a flowchart of a method for testing a face recognition performance according to an embodiment of the present invention, as shown in fig. 6, the method includes the following steps:
step S602, the static face detection image is subjected to image quality detection, and is analyzed and classified, and the unqualified condition of the static face detection image comprises the following steps: the resolution ratio does not satisfy: resolution is lower than 640 x 480 pixels; incomplete face: occlusion exists in five sense organs of the face in the image; interpupillary distance is not satisfied: the interpupillary distance is less than 60 pixels; blurring the image; the facial expression is not satisfied: expressions with closed eyes or open mouth; the angle does not satisfy: the absolute value of the rotation angle, pitch angle or tilt angle is greater than 20 °; the illumination does not satisfy: high image contrast, shadows on the face, image overexposure, or image underexposure.
Analyzing the test log according to the image quality detection, and acquiring the face detection counting index; wherein the image quality detection comprises at least one of: resolution, face integrity, face interpupillary distance, image definition, face expression, face angle and image contrast; the face detection performance test table is shown in table 3:
table 3 face detection performance test table
Through the step S602, the image quality of the static face detection image is detected, and the face detection count index is obtained, so that the performance and accuracy of the face detection algorithm are judged, the reliability of the face detection threshold setting range is judged, and the comprehensiveness of the face recognition performance test is further improved.
In one embodiment, a method for testing face recognition performance is provided, which further includes the following steps:
step S702, sending the face detection counting index to a terminal; the terminal displays the face detection counting index in a chart form such as tables 1 to 3, so that testers can timely master and analyze specific results of the face recognition performance test.
It should be understood that, although the steps in the flowcharts of fig. 2 to 6 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.
An embodiment of the present invention is described in detail below with reference to an actual application scenario, and fig. 7 is a flowchart six of a face recognition performance testing method according to an embodiment of the present invention, where specific implementation steps of the face recognition performance testing method are shown in fig. 7.
Step S802, acquiring video stream images and setting human face detection times; and carrying out face detection on the video stream image according to a face detection algorithm.
Step S804, judging whether a human face is detected; if not, saving the picture and carrying out picture quality detection, including: whether the multiple detections are consistent, the false detection rate of face detection, the excessive angle probability, the illumination influence probability, the shielding probability, the distance influence probability, the picture missing probability and the like, and if the face is detected, the face detection passing rate is obtained.
Step S806, setting the number of times of live body detection and a live body threshold value, and calling a live body detection algorithm; or if the living body is not detected, storing the non-living body picture and carrying out multiple detections, analyzing the non-living body picture according to the center algorithm model, the crop algorithm model and the mouse algorithm model, respectively judging whether the center part of the face is correct, whether the mouth is correct, whether any two parts of the face are symmetrical, and determining the face as the non-living body if any one of the three parts is not correct; if a living body is detected, a living body passage rate is acquired.
Step S808, setting comparison times and a comparison threshold value, and registering a characteristic database; detecting according to a face comparison algorithm; judging whether the detection result is greater than a comparison threshold value, and if not, acquiring the rejection rate of the face comparison; and if the comparison threshold is larger than the comparison threshold, acquiring the face comparison passing rate.
In this embodiment, a device for testing face recognition performance is provided, and fig. 8 is a block diagram of a structure of the device for testing face recognition performance according to the embodiment of the present invention, as shown in fig. 8, the system includes:
a face detection module 82, configured to perform face detection on the acquired real-time streaming image according to a face detection algorithm;
a saving module 84, configured to save the real-time streaming image as a static face detection image when it is determined that the face detection fails according to the face detection threshold; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the same face detection rate according to the result of the static face detection;
an output module 86, configured to stop the face detection when the face detection is performed multiple times and the total number of face detections is greater than or equal to a preset number of face detections, and output the same rate of face detection and face detection data to a test log;
and an obtaining module 88, configured to obtain a face detection count index according to the static face detection image and the test log, where the face detection count index represents a face recognition performance.
Through the embodiment, the face detection module 82 performs face detection on the acquired real-time stream image according to a face detection algorithm, the storage module 84 stores the real-time stream image which fails in face detection as a static image, and detects and confirms the static image again, so that the reason why the face is not detected is conveniently and quickly analyzed; meanwhile, the acquisition module 88 acquires a face detection counting index according to the face detection data for representing the face recognition performance, so that the face recognition performance is automatically tested, and the problem of low testing efficiency in the face recognition performance test is solved.
In an embodiment, a device for testing face recognition performance is provided, fig. 9 is a block diagram of a structure of a device for testing face recognition performance according to an embodiment of the present invention, as shown in fig. 9, the device further includes a living body detection module 92;
the live body detection module 92 performs live body detection on the real-time streaming image according to a live body detection algorithm under the condition that the face detection passes;
the saving module 84 saves the live streaming image as a static live detection image when the live detection is judged not to pass according to the live detection threshold; performing static biopsy on the static biopsy image according to the biopsy algorithm, and acquiring the biopsy identity rate according to the result of the static biopsy;
the output module 86 performs the biopsy a plurality of times, stops the biopsy and outputs the same rate of biopsy and biopsy data to the test log if the total number of biopsies is greater than or equal to the preset number of biopsies;
the obtaining module 88 obtains a living body detection counting index according to the static living body detection image and the test log, and represents the face recognition performance according to the living body detection counting index and the face detection counting index.
In an embodiment, a device for testing a face recognition performance is provided, and fig. 10 is a block diagram of a structure of the device for testing a face recognition performance according to the embodiment of the present invention, as shown in fig. 10, the device further includes a face comparison module 102;
the face comparison module 102 performs face comparison on the real-time streaming image according to a face comparison algorithm when the living body detection passes;
the saving module 84 saves the real-time streaming image as a static face comparison image under the condition that the face comparison is judged not to pass according to the face comparison threshold; performing static face comparison on the static face comparison image according to the face comparison algorithm, and acquiring the face comparison identity rate according to the static face comparison result;
the output module 86 performs the face comparison for a plurality of times, stops the face comparison when the total number of the face comparison is greater than or equal to the preset number of the face comparison, and outputs the same face comparison rate and face comparison data to the test log;
the obtaining module 88 obtains a face comparison count index according to the static face comparison image and the test log, and represents the face recognition performance according to the face comparison count index, the living body detection count index and the face detection count index.
In one embodiment, the obtaining module 88 is further configured to perform a biopsy analysis on the static biopsy image and obtain the biopsy count indicator according to the biopsy analysis and the test log; wherein the in vivo detection assay comprises: illumination analysis, paper analysis, and prosthesis analysis.
In one embodiment, the obtaining module 88 is further configured to perform image quality detection on the static face detection image, and the obtaining module 88 analyzes the test log according to the image quality detection and obtains the face detection count index; wherein the image quality detection comprises at least one of: resolution, face integrity, face interpupillary distance, image sharpness, face expression, face angle, and image contrast.
In one embodiment, the apparatus further comprises a display module; the display module sends the face detection count index to the terminal 12; wherein, the terminal 12 displays the face detection counting index in the form of a chart.
For the specific limitation of the face recognition performance testing apparatus, reference may be made to the above limitation of the face recognition performance testing method, and details are not described herein again. All or part of the modules in the face recognition performance testing device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, and the computer device may be a server, and fig. 11 is a structural diagram of the inside of a computer device according to an embodiment of the present invention, as shown in fig. 11. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing test log data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a face recognition performance testing method.
Those skilled in the art will appreciate that the architecture shown in fig. 11 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the steps in the face comparison method provided in the foregoing embodiments are implemented.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps in the face comparison method provided in the foregoing embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for testing face recognition performance is characterized by comprising the following steps:
carrying out face detection on the acquired real-time stream image according to a face detection algorithm;
under the condition that the human face detection is judged not to pass according to a human face detection threshold value, the real-time streaming image is stored as a static human face detection image; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the face detection identity rate according to the result of the static face detection;
performing the face detection for multiple times, and outputting the face detection same rate and face detection data to a test log under the condition that the total number of the face detection is greater than or equal to the preset number of the face detection;
and acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance.
2. The method of claim 1, wherein after the face detection of the acquired real-time streaming image according to the face detection algorithm, the method further comprises:
performing living body detection on the real-time stream image according to a living body detection algorithm under the condition that the face detection is passed;
saving the real-time stream image as a static biopsy image under the condition that the biopsy is judged not to pass according to a biopsy threshold; performing static living body detection on the static living body detection image according to the living body detection algorithm, and acquiring the same rate of the living body detection according to the result of the static living body detection;
performing the live body detection a plurality of times, and outputting the same rate of live body detection and live body detection data to the test log in the case that the total number of the live body detections is greater than or equal to the preset number of live body detections;
and acquiring a living body detection counting index according to the static living body detection image and the test log, and representing the face identification performance according to the living body detection counting index and the face detection counting index.
3. The method of claim 2, wherein after the live-detection of the live-streaming image according to a live-detection algorithm, the method further comprises:
under the condition that the living body detection is passed, carrying out face comparison on the real-time stream image according to a face comparison algorithm;
under the condition that the face comparison is not passed according to a face comparison threshold value, storing the real-time streaming image as a static face comparison image; performing static face comparison on the static face comparison image according to the face comparison algorithm, and acquiring the face comparison identity rate according to the static face comparison result;
performing face comparison for multiple times, and outputting the same face comparison rate and face comparison data to the test log under the condition that the total number of face comparison is greater than or equal to the preset number of face comparison;
and acquiring a face comparison counting index according to the static face comparison image and the test log, and representing the face recognition performance according to the face comparison counting index, the living body detection counting index and the face detection counting index.
4. The method of claim 2, wherein obtaining a liveness count indicator from the static liveness image and the test log comprises:
performing in-vivo detection analysis on the static in-vivo detection image, and acquiring the in-vivo detection counting index according to the in-vivo detection analysis and the test log; wherein the in vivo detection assay comprises: illumination analysis, paper analysis, and prosthesis analysis.
5. The method of claim 1, wherein obtaining a detection count indicator from the static face detection image and the test log comprises:
performing image quality detection on the static face detection image, analyzing the test log according to the image quality detection, and acquiring the face detection counting index; wherein the image quality detection comprises at least one of: resolution, face integrity, face interpupillary distance, image sharpness, face expression, face angle, and image contrast.
6. The method according to any one of claims 1 to 5, wherein after obtaining a face detection count indicator from the static face detection image and the test log, the method further comprises:
sending the face detection counting index to a terminal; and the terminal displays the face detection counting index in a form of a chart.
7. An apparatus for testing face recognition performance, the apparatus comprising:
the face detection module is used for carrying out face detection on the acquired real-time stream image according to a face detection algorithm;
the storage module is used for storing the real-time streaming image as a static face detection image under the condition that the face detection is judged not to pass according to a face detection threshold value; performing static face detection on the static face detection image according to the face detection algorithm, and acquiring the face detection identity rate according to the result of the static face detection;
the output module is used for outputting the face detection identical rate and face detection data to a test log when the face detection is carried out for a plurality of times and the total number of times of the face detection is greater than or equal to the preset number of times of the face detection;
and the acquisition module is used for acquiring a face detection counting index according to the static face detection image and the test log, wherein the face detection counting index represents the face recognition performance.
8. The device of claim 7, further comprising a liveness detection module;
the living body detection module performs living body detection on the real-time stream image according to a living body detection algorithm under the condition that the face detection is passed;
the storage module stores the real-time stream image as a static living body detection image under the condition that the living body detection is judged not to pass according to a living body detection threshold; performing static living body detection on the static living body detection image according to the living body detection algorithm, and acquiring the same rate of the living body detection according to the result of the static living body detection;
the output module performs the living body detection for a plurality of times, and outputs the same living body detection rate and living body detection data to the test log when the total number of the living body detections is greater than or equal to the preset number of living body detections;
the acquisition module acquires a living body detection counting index according to the static living body detection image and the test log, and represents the face recognition performance according to the living body detection counting index and the face detection counting index.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 6 are implemented when the computer program is executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911418325.0A CN111241945A (en) | 2019-12-31 | 2019-12-31 | Method and device for testing face recognition performance, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911418325.0A CN111241945A (en) | 2019-12-31 | 2019-12-31 | Method and device for testing face recognition performance, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111241945A true CN111241945A (en) | 2020-06-05 |
Family
ID=70865504
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911418325.0A Pending CN111241945A (en) | 2019-12-31 | 2019-12-31 | Method and device for testing face recognition performance, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111241945A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783663A (en) * | 2020-06-30 | 2020-10-16 | 公安部第三研究所 | Algorithm evaluation system and detection method for performance detection of human evidence verification equipment |
| CN115601799A (en) * | 2022-09-09 | 2023-01-13 | 广州市盛通建设工程质量检测有限公司（Cn） | Evaluation method, system, equipment and storage medium based on face recognition |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101159016A (en) * | 2007-11-26 | 2008-04-09 | 清华大学 | Living body detecting method and system based on human face physiologic moving |
| CN109063617A (en) * | 2018-07-24 | 2018-12-21 | 高新兴科技集团股份有限公司 | A kind of analysis method and tool of face recognition algorithms |
| CN110162462A (en) * | 2019-04-16 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Test method, system and the computer equipment of face identification system based on scene |
-
2019
- 2019-12-31 CN CN201911418325.0A patent/CN111241945A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101159016A (en) * | 2007-11-26 | 2008-04-09 | 清华大学 | Living body detecting method and system based on human face physiologic moving |
| CN109063617A (en) * | 2018-07-24 | 2018-12-21 | 高新兴科技集团股份有限公司 | A kind of analysis method and tool of face recognition algorithms |
| CN110162462A (en) * | 2019-04-16 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Test method, system and the computer equipment of face identification system based on scene |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783663A (en) * | 2020-06-30 | 2020-10-16 | 公安部第三研究所 | Algorithm evaluation system and detection method for performance detection of human evidence verification equipment |
| CN115601799A (en) * | 2022-09-09 | 2023-01-13 | 广州市盛通建设工程质量检测有限公司（Cn） | Evaluation method, system, equipment and storage medium based on face recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110443110B (en) | Face recognition method, device, terminal and storage medium based on multipath camera shooting | |
| US11093773B2 (en) | Liveness detection method, apparatus and computer-readable storage medium | |
| CN110490076B (en) | Living body detection method, living body detection device, computer equipment and storage medium | |
| WO2021068322A1 (en) | Training method and apparatus for living body detection model, computer device, and storage medium | |
| CN110659397B (en) | Behavior detection method and device, electronic equipment and storage medium | |
| CN112801057B (en) | Image processing method, image processing device, computer equipment and storage medium | |
| CN109299658B (en) | Face detection method, face image rendering device and storage medium | |
| CN110969045B (en) | Behavior detection method and device, electronic equipment and storage medium | |
| CN113920309B (en) | Image detection method, image detection device, medical image processing equipment and storage medium | |
| CN111191521B (en) | Face living body detection method and device, computer equipment and storage medium | |
| CN111144285B (en) | Fat and thin degree identification method, device, equipment and medium | |
| CN111444555B (en) | Temperature measurement information display method and device and terminal equipment | |
| CN113516661A (en) | Defect detection method and device based on feature fusion | |
| CN111241945A (en) | Method and device for testing face recognition performance, computer equipment and storage medium | |
| CN112884782A (en) | Biological object segmentation method, apparatus, computer device and storage medium | |
| CN113255516A (en) | Living body detection method and device and electronic equipment | |
| CN113971841A (en) | Living body detection method and device, computer equipment and storage medium | |
| CN110163183B (en) | Target detection algorithm evaluation method and device, computer equipment and storage medium | |
| CN112308035A (en) | Image detection method, image detection device, computer equipment and storage medium | |
| CN112241705A (en) | Target detection model training method and target detection method based on classification regression | |
| CN113553990B (en) | Method and device for tracking and identifying multiple faces, computer equipment and storage medium | |
| CN112465037B (en) | Target detection method, device, computer equipment and storage medium | |
| CN114332981A (en) | Face living body detection method and device, electronic equipment and storage medium | |
| Kuznetsov et al. | Cross-Database Liveness Detection: Insights from Comparative Biometric Analysis | |
| CN113240043A (en) | False identification method, device and equipment based on multi-picture difference and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20231208 |
|
| AD01 | Patent right deemed abandoned |