CN111131693A - Face image enhancement method based on multi-exposure face detection - Google Patents
Face image enhancement method based on multi-exposure face detection Download PDFInfo
- Publication number
- CN111131693A CN111131693A CN201911084025.3A CN201911084025A CN111131693A CN 111131693 A CN111131693 A CN 111131693A CN 201911084025 A CN201911084025 A CN 201911084025A CN 111131693 A CN111131693 A CN 111131693A
- Authority
- CN
- China
- Prior art keywords
- exposure
- human face
- algorithm
- face
- dynamic range
- 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.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/61—Control of cameras or camera modules based on recognised objects
- H04N23/611—Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/73—Circuitry for compensating brightness variation in the scene by influencing the exposure time
Abstract
The invention discloses a method for enhancing a face image based on multi-exposure face detection, which comprises the following steps: carrying out face detection on the image frame input into the face detection model and acquiring the face position; segmenting the image into M x N regions; if the face position is detected, different weights are directly set for the M-N areas according to a designed algorithm; if the human face is not detected, judging the dynamic range of the current scene, and determining whether to switch the exposure parameters according to the dynamic range; if the dynamic range is smaller than a preset threshold value, exposure parameters do not need to be switched, the current scene is judged to have no human face, and the weight of the M-N block area is set as a default weight; if the dynamic range is larger than the preset threshold value, different exposure quantities and exposure parameters are switched according to a preset multi-exposure switching algorithm, and the human face is detected under different exposure parameters.
Description
Technical Field
The invention belongs to a face image enhancement method based on multi-exposure face detection, relates to the field of a camera 3A and machine vision, and particularly relates to a method for enhancing a face region by improving an automatic exposure algorithm in combination with machine vision face detection.
Background
With the rise of machine vision, the face detection algorithm also achieves a nearly perfect effect on the existing data set, however, the existing data set has the characteristics of clear face and reasonable exposure; the existing face detection algorithm with the strongest comprehensive capability also has a greatly reduced detection effect in a severe illumination environment, such as environments with backlight, low light and the like. The machine learning field face detection is based on human face characteristics and matched with a corresponding classification algorithm learning model, so that whether a face exists in an input image frame or not is judged, and the approximate position of the face in the image can be obtained; the accurate position of the face in the image can be output by face detection in the deep learning, and the deep learning also has higher accuracy, but the deep learning needs higher hardware requirement and longer detection time.
The traditional exposure algorithm generally counts a global brightness mean value, or sets a weighted brightness mean value with fixed weight for a default region of interest, and calculates exposure configuration exposure parameters based on the global brightness mean value or the weighted brightness mean value. Under the scene that the light is uniform or the whole illumination is not uniform, but the illumination of the default region of interest is uniform and the face is just in the default region of interest, the traditional exposure algorithm has a good exposure effect on the face, but under the special illumination condition, for example, the illumination is not uniform, but the face is just not in the default region of interest, the phenomenon that the face region is abnormally exposed can occur, so that the negative influence on the subsequent machine vision is realized, and the direct human eye visual effect can also be influenced.
The existing system method for carrying out intelligent exposure based on human faces is characterized in that a single-frame image is input, the human faces are detected on the single-frame image, an exposure weight table is modified according to the position information of the human faces, and different exposure amounts and exposure parameters are further obtained.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method for enhancing a human face image based on multi-exposure human face detection.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a method for enhancing face images based on multi-exposure face detection comprises the following steps:
carrying out face detection on the image frame input into the face detection model and acquiring the face position;
segmenting the image into M x N regions; if the face position is detected, different weights are directly set for the M-N areas according to a designed algorithm;
if the human face is not detected, judging the dynamic range of the current scene, and determining whether to switch the exposure parameters according to the dynamic range;
if the dynamic range is smaller than a preset threshold value, exposure parameters do not need to be switched, the current scene is judged to have no human face, and the weight of the M-N block area is set as a default weight;
if the dynamic range is larger than a preset threshold value, different exposure quantities and exposure parameters are switched according to a preset multi-exposure switching algorithm, and the human face is detected under different exposure parameters;
if the face is detected, setting different weights for the M x N areas according to a designed algorithm;
if the face is not detected, judging that the current scene has no face, namely, not deeply detecting the face downwards, and setting the weight of the M x N block regions as a default weight;
obtaining the weighted statistical parameters required by the automatic exposure algorithm through different weights of the M-N block areas;
obtaining the exposure required by automatic exposure according to the obtained statistical parameters;
the exposure parameters are distributed according to the preset exposure route, so that an image for enhancing the face area based on the face detection is obtained.
Preferably, the method further comprises the following steps:
training a face detection model in advance, selecting an automatic exposure algorithm, and performing a multi-exposure switching algorithm according to the brightness information of the current scene as a condition.
Preferably, the face detection of the image frame input to the face detection model and the face position acquisition specifically include: and selecting a human face detection model, judging whether a human face exists in the input image frame according to the human face characteristics, and outputting the position of the human face.
Preferably, the obtaining of the weighted statistical parameters required by the automatic exposure algorithm through different weights of the M × N block regions specifically includes:
the automatic exposure algorithm calculates proper exposure according to the current scene brightness, so that the image can correctly describe the scene brightness information, and the simple principle is as follows:
IErepresenting the target brightness of the image, IKThe image brightness corresponding to the current exposure parameter is defined, and the Bias is the allowable brightness deviation; k represents the current exposure parameter, and L represents the actual brightness of the current scene;
an exposure weight table, which divides the image into M × N regions, each region corresponds to different weights, and a matrix composed of M × N weights corresponding to M × N block regions is the exposure weight table, which is used for weighting different regions, so as to obtain weighted statistical information, namely I in formula (1)k；
The exposure route is used for distributing specific exposure parameters according to exposure parameters such as an exposure distribution diaphragm, exposure time, gain and the like;
the dynamic range is used for representing the brightness and darkness phase difference level of the scene, and is specifically calculated as follows:
where DR represents dynamic range, unit: dB, i _ max and i _ min represent the maximum value and the minimum value of the brightness in the scene respectively;
and the dynamic range threshold is used for judging the high and low of the scene dynamic range so as to decide whether to switch to a multi-exposure algorithm, wherein the dynamic range threshold is higher when being larger than the threshold and is lower when being smaller than the threshold.
Preferably, if the dynamic range is greater than a preset threshold, switching different exposure amounts and exposure parameters according to a preset multi-exposure switching algorithm, and detecting a human face under different exposure parameters includes:
when the current image frame can not effectively detect the human face, the exposure is switched according to a multi-exposure algorithm, the multi-exposure algorithm can be set by a user, the exposure can be increased or decreased according to the equal ratio or the equal quantity of the current exposure until the multi-exposure can meet the requirement of the whole dynamic range, the switching step length between different exposures needs the dynamic range of the current scene, and the video output real-time performance is comprehensively considered, and the simple principle of enhancing the human face detection rate by a multi-exposure image sequence is as follows:
wherein p isiDetecting rates of the face detection model on image sequences with different exposure quantities;
and the multi-exposure image sequence is used as the input of the human face detection according to the image sequences with different exposure output by the multi-exposure algorithm, so that the human face detection rate under the same scene is enhanced.
Preferably, a human face detection model with higher overall score of detection rate and detection time is pre-trained for the algorithm based on hardware conditions; dividing the image into reasonable M x N areas;
setting a default region of interest for a default exposure algorithm, wherein if a vehicle-mounted camera needs to set the middle lower part as the region of interest, and increasing the weight of the region of interest, the algorithm uses the default exposure weight table when the initialization and the human face cannot be detected;
under the condition that the face detection algorithm works, setting different weights for M-N areas for modifying an exposure weight table according to position information output by the detection algorithm;
counting statistics required by automatic exposure according to an exposure weight table to obtain the corresponding exposure amount in the scene;
distributing exposure parameters according to the exposure route and applying the parameters to the board;
when the face detection of the current frame fails, judging the dynamic range of the current scene;
if the dynamic range is smaller than a preset threshold value, the current scene is judged to have no human face, and a default exposure weight table is switched to prepare for subsequent human face detection.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The present invention will be described in detail below with reference to the accompanying drawings so that the above advantages of the present invention will be more apparent. Wherein the content of the first and second substances,
FIG. 1 is a flow chart of a method for enhancing a face image based on multi-exposure face detection according to the present invention.
Detailed Description
The following detailed description of the embodiments of the present invention will be provided with reference to the drawings and examples, so that how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. It should be noted that, as long as there is no conflict, the embodiments and the features of the embodiments of the present invention may be combined with each other, and the technical solutions formed are within the scope of the present invention.
The first objective of the present invention is to enhance the image quality of the face area by intelligently exposing the camera in conjunction with face detection.
The second objective of the invention is to detect the human face by using the image sequences of different exposure and exposure parameters of a plurality of frames under the condition that the detection of the human face by using a single frame image is invalid, thereby increasing the human face detection rate.
A third object of the present invention is to provide an intelligent exposure algorithm with low complexity so that it can run in real time on an embedded platform.
In order to achieve the above object, the present invention provides a method for performing face detection on a multi-exposure image sequence, and modifying an exposure weight table according to a position obtained by the face detection to obtain different exposure amounts and subsequent exposure parameters for performing intelligent exposure, thereby enhancing the image quality of a face region.
The human face detection model is used for judging whether a human face exists in an input image frame according to the human face characteristics and outputting the position of the human face;
the automatic exposure algorithm calculates proper exposure according to the current scene brightness, so that the image can correctly describe the scene brightness information, and the simple principle is as follows:
IErepresenting the target brightness of the image, IKThe image brightness corresponding to the current exposure parameter is defined, and the Bias is the allowable brightness deviation; k represents the current exposure parameter, and L represents the actual brightness of the current scene;
an exposure weight table, which divides the image into M × N regions, each region corresponds to different weights, and a matrix composed of M × N weights corresponding to M × N block regions is the exposure weight table, which is used for weighting different regions, so as to obtain weighted statistical information, namely I in formula (1)k；
The exposure route is used for distributing specific exposure parameters according to exposure parameters such as an exposure distribution diaphragm, exposure time, gain and the like;
the dynamic range is used for representing the brightness and darkness phase difference level of the scene, and is specifically calculated as follows:
where DR represents dynamic range, unit: dB, i _ max and i _ min represent the maximum value and the minimum value of the brightness in the scene respectively;
the dynamic range threshold is used for judging the height of the scene dynamic range so as to determine whether to switch to a multi-exposure algorithm, wherein the height is higher when the dynamic range threshold is larger than the threshold and the height is lower when the dynamic range threshold is smaller than the threshold;
the multi-exposure algorithm is characterized in that when a face cannot be effectively detected in a current image frame, exposure is switched according to the multi-exposure algorithm, the multi-exposure algorithm can be set by a user, the exposure can be increased or decreased according to the equal proportion or equal quantity of the current exposure until the multi-exposure can meet the requirement of the whole dynamic range, the switching step length between different exposures needs the dynamic range of the current scene, and the video output real-time performance is comprehensively considered, and the simple principle of enhancing the face detection rate by a multi-exposure image sequence is as follows
Wherein p isiDetecting rates of the face detection model on image sequences with different exposure quantities;
and the multi-exposure image sequence is used as the input of the human face detection according to the image sequences with different exposure output by the multi-exposure algorithm, so that the human face detection rate under the same scene is enhanced.
For the above modules to implement the basic module and the specific function of the present invention, the working logic and the working sequence between different modules will be described as follows:
firstly, a human face detection model with high overall score of detection rate and detection time is pre-trained for the algorithm based on hardware conditions;
dividing the image into reasonable M x N areas;
setting a default region of interest for a default exposure algorithm, wherein the step needs to be determined according to an application scene, for example, a vehicle-mounted camera needs to set the middle lower part as the region of interest and increase the weight of the region of interest, and the algorithm uses the default exposure weight table when the initialization and the human face cannot be detected;
under the condition that the face detection algorithm works, setting different weights for M-N areas for modifying an exposure weight table according to position information output by the detection algorithm;
counting statistics required by automatic exposure according to an exposure weight table to obtain the corresponding exposure amount in the scene;
distributing exposure parameters according to the exposure route and applying the parameters to the board;
when the face detection of the current frame fails, judging the dynamic range of the current scene;
if the dynamic range is smaller than a preset threshold value, judging that the current scene has no human face, and switching to a default exposure weight table to prepare for subsequent human face detection;
if the dynamic range is higher than the threshold value, the following steps are carried out:
setting different exposure switching routes according to the current exposure and a multi-exposure algorithm;
detecting a human face on a multi-exposure image sequence, modifying an exposure weight table according to position information output by a detection algorithm after the human face is detected, setting different weights for M x N areas, counting exposure information according to the exposure weight table to obtain corresponding exposure amount in the scene, distributing exposure parameters according to an exposure route, applying the exposure parameters to a board, and if the human face is not detected in the multi-exposure image sequence, judging that the current scene has no human face, switching to a default exposure weight table to prepare for subsequent human face detection.
Specifically, a method for enhancing a face image based on multi-exposure face detection comprises the following steps:
carrying out face detection on the image frame input into the face detection model and acquiring the face position;
segmenting the image into M x N regions; if the face position is detected, different weights are directly set for the M-N areas according to a designed algorithm;
if the human face is not detected, judging the dynamic range of the current scene, and determining whether to switch the exposure parameters according to the dynamic range;
if the dynamic range is smaller than a preset threshold value, exposure parameters do not need to be switched, the current scene is judged to have no human face, and the weight of the M-N block area is set as a default weight;
if the dynamic range is larger than a preset threshold value, different exposure quantities and exposure parameters are switched according to a preset multi-exposure switching algorithm, and the human face is detected under different exposure parameters;
if the face is detected, setting different weights for the M x N areas according to a designed algorithm;
if the face is not detected, judging that the current scene has no face, namely, not deeply detecting the face downwards, and setting the weight of the M x N block regions as a default weight;
obtaining the weighted statistical parameters required by the automatic exposure algorithm through different weights of the M-N block areas;
obtaining the exposure required by automatic exposure according to the obtained statistical parameters;
the exposure parameters are distributed according to the preset exposure route, so that an image for enhancing the face area based on the face detection is obtained.
Preferably, the method further comprises the following steps:
training a face detection model in advance, selecting an automatic exposure algorithm, and performing a multi-exposure switching algorithm according to the brightness information of the current scene as a condition.
Preferably, the face detection of the image frame input to the face detection model and the face position acquisition specifically include: and selecting a human face detection model, judging whether a human face exists in the input image frame according to the human face characteristics, and outputting the position of the human face.
Preferably, the obtaining of the weighted statistical parameters required by the automatic exposure algorithm through different weights of the M × N block regions specifically includes:
the automatic exposure algorithm calculates proper exposure according to the current scene brightness, so that the image can correctly describe the scene brightness information, and the simple principle is as follows:
IErepresenting the target brightness of the image, IKThe image brightness corresponding to the current exposure parameter is defined, and the Bias is the allowable brightness deviation; k represents the current exposure parameter, and L represents the actual brightness of the current scene;
an exposure weight table, which divides the image into M × N regions, each region corresponds to different weights, and a matrix composed of M × N weights corresponding to M × N block regions is the exposure weight table, which is used for weighting different regions, so as to obtain weighted statistical information, namely I in formula (1)k；
The exposure route is used for distributing specific exposure parameters according to exposure parameters such as an exposure distribution diaphragm, exposure time, gain and the like;
the dynamic range is used for representing the brightness and darkness phase difference level of the scene, and is specifically calculated as follows:
where DR represents dynamic range, unit: dB, i _ max and i _ min represent the maximum value and the minimum value of the brightness in the scene respectively;
and the dynamic range threshold is used for judging the high and low of the scene dynamic range so as to decide whether to switch to a multi-exposure algorithm, wherein the dynamic range threshold is higher when being larger than the threshold and is lower when being smaller than the threshold.
Preferably, if the dynamic range is greater than a preset threshold, switching different exposure amounts and exposure parameters according to a preset multi-exposure switching algorithm, and detecting a human face under different exposure parameters includes:
when the current image frame can not effectively detect the human face, the exposure is switched according to a multi-exposure algorithm, the multi-exposure algorithm can be set by a user, the exposure can be increased or decreased according to the equal ratio or the equal quantity of the current exposure until the multi-exposure can meet the requirement of the whole dynamic range, the switching step length between different exposures needs the dynamic range of the current scene, and the video output real-time performance is comprehensively considered, and the simple principle of enhancing the human face detection rate by a multi-exposure image sequence is as follows:
wherein p isiDetecting rates of the face detection model on image sequences with different exposure quantities;
and the multi-exposure image sequence is used as the input of the human face detection according to the image sequences with different exposure output by the multi-exposure algorithm, so that the human face detection rate under the same scene is enhanced.
Preferably, a human face detection model with higher overall score of detection rate and detection time is pre-trained for the algorithm based on hardware conditions; dividing the image into reasonable M x N areas;
setting a default region of interest for a default exposure algorithm, wherein if a vehicle-mounted camera needs to set the middle lower part as the region of interest, and increasing the weight of the region of interest, the algorithm uses the default exposure weight table when the initialization and the human face cannot be detected;
under the condition that the face detection algorithm works, setting different weights for M-N areas for modifying an exposure weight table according to position information output by the detection algorithm;
counting statistics required by automatic exposure according to an exposure weight table to obtain the corresponding exposure amount in the scene;
distributing exposure parameters according to the exposure route and applying the parameters to the board;
when the face detection of the current frame fails, judging the dynamic range of the current scene;
if the dynamic range is smaller than a preset threshold value, the current scene is judged to have no human face, and a default exposure weight table is switched to prepare for subsequent human face detection.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A method for enhancing a face image based on multi-exposure face detection is characterized by comprising the following steps:
carrying out face detection on the image frame input into the face detection model and acquiring the face position;
segmenting the image into M x N regions; if the face position is detected, different weights are directly set for the M-N areas according to a designed algorithm;
if the human face is not detected, judging the dynamic range of the current scene, and determining whether to switch the exposure parameters according to the dynamic range;
if the dynamic range is smaller than a preset threshold value, exposure parameters do not need to be switched, the current scene is judged to have no human face, and the weight of the M-N block area is set as a default weight;
if the dynamic range is larger than a preset threshold value, different exposure quantities and exposure parameters are switched according to a preset multi-exposure switching algorithm, and the human face is detected under different exposure parameters;
if the face is detected, setting different weights for the M x N areas according to a designed algorithm;
if the face is not detected, judging that the current scene has no face, namely, not deeply detecting the face downwards, and setting the weight of the M x N block regions as a default weight;
obtaining the weighted statistical parameters required by the automatic exposure algorithm through different weights of the M-N block areas;
obtaining the exposure required by automatic exposure according to the obtained statistical parameters;
the exposure parameters are distributed according to the preset exposure route, so that an image for enhancing the face area based on the face detection is obtained.
2. The method for enhancing a human face image based on multi-exposure human face detection according to claim 1, further comprising:
training a face detection model in advance, selecting an automatic exposure algorithm, and performing a multi-exposure switching algorithm according to the brightness information of the current scene as a condition.
3. The method for enhancing a human face image based on multi-exposure human face detection as claimed in claim 2, wherein the human face detection and the obtaining of the human face position are performed on the image frame input to the human face detection model, and specifically comprises: and selecting a human face detection model, judging whether a human face exists in the input image frame according to the human face characteristics, and outputting the position of the human face.
4. The method for enhancing a human face image based on multi-exposure human face detection according to claim 1, wherein the weighted statistical parameters required by the automatic exposure algorithm are obtained by different weights of M × N regions, and specifically comprises:
the automatic exposure algorithm calculates proper exposure according to the current scene brightness, so that the image can correctly describe the scene brightness information, and the simple principle is as follows:
IErepresenting the target brightness of the image, IKThe image brightness corresponding to the current exposure parameter is defined, and the Bias is the allowable brightness deviation; k represents the current exposure parameter, and L represents the actual brightness of the current scene;
an exposure weight table for dividing the image into M × N regions, each region corresponding to different weights, and M × N weights corresponding to M × N regionsThe matrix is an exposure weight table, which is used to weight different areas to obtain weighted statistical information, i.e. I in formula (1)k；
The exposure route is used for distributing specific exposure parameters according to exposure parameters such as an exposure distribution diaphragm, exposure time, gain and the like;
the dynamic range is used for representing the brightness and darkness phase difference level of the scene, and is specifically calculated as follows:
where DR represents dynamic range, unit: dB, i _ max and i _ min represent the maximum value and the minimum value of the brightness in the scene respectively;
and the dynamic range threshold is used for judging the high and low of the scene dynamic range so as to decide whether to switch to a multi-exposure algorithm, wherein the dynamic range threshold is higher when being larger than the threshold and is lower when being smaller than the threshold.
5. The method of claim 1, wherein if the dynamic range is larger than a preset threshold, the method switches different exposure amounts and exposure parameters according to a preset multi-exposure switching algorithm, and detects the face under different exposure parameters, comprising:
when the current image frame can not effectively detect the human face, the exposure is switched according to a multi-exposure algorithm, the multi-exposure algorithm can be set by a user, the exposure can be increased or decreased according to the equal ratio or the equal quantity of the current exposure until the multi-exposure can meet the requirement of the whole dynamic range, the switching step length between different exposures needs the dynamic range of the current scene, and the video output real-time performance is comprehensively considered, and the simple principle of enhancing the human face detection rate by a multi-exposure image sequence is as follows:
wherein p isiImage sequence for face detection model at different exposureDetection rate on the columns;
and the multi-exposure image sequence is used as the input of the human face detection according to the image sequences with different exposure output by the multi-exposure algorithm, so that the human face detection rate under the same scene is enhanced.
6. The method for face image enhancement based on multi-exposure face detection according to claim 1,
pre-training a face detection model with higher overall score of detection rate and detection time for the algorithm based on hardware conditions; dividing the image into reasonable M x N areas;
setting a default region of interest for a default exposure algorithm, wherein if a vehicle-mounted camera needs to set the middle lower part as the region of interest, and increasing the weight of the region of interest, the algorithm uses the default exposure weight table when the initialization and the human face cannot be detected;
under the condition that the face detection algorithm works, setting different weights for M-N areas for modifying an exposure weight table according to position information output by the detection algorithm;
counting statistics required by automatic exposure according to an exposure weight table to obtain the corresponding exposure amount in the scene;
distributing exposure parameters according to the exposure route and applying the parameters to the board;
when the face detection of the current frame fails, judging the dynamic range of the current scene;
if the dynamic range is smaller than a preset threshold value, the current scene is judged to have no human face, and a default exposure weight table is switched to prepare for subsequent human face detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911084025.3A CN111131693B (en) | 2019-11-07 | 2019-11-07 | Face image enhancement method based on multi-exposure face detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911084025.3A CN111131693B (en) | 2019-11-07 | 2019-11-07 | Face image enhancement method based on multi-exposure face detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111131693A true CN111131693A (en) | 2020-05-08 |
| CN111131693B CN111131693B (en) | 2021-07-30 |
Family
ID=70495748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911084025.3A Active CN111131693B (en) | 2019-11-07 | 2019-11-07 | Face image enhancement method based on multi-exposure face detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111131693B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112511749A (en) * | 2020-11-30 | 2021-03-16 | 上海摩象网络科技有限公司 | Automatic exposure control method and device for target object and electronic equipment |
| CN112839183A (en) * | 2020-12-28 | 2021-05-25 | 杭州宇泛智能科技有限公司 | Environment self-adaptive face image recognition method |
| CN112866773A (en) * | 2020-08-21 | 2021-05-28 | 海信视像科技股份有限公司 | Display device and camera tracking method in multi-person scene |
| CN112911146A (en) * | 2021-01-27 | 2021-06-04 | 杭州寰宇微视科技有限公司 | Intelligent dimming method based on human face |
| CN114727024A (en) * | 2021-01-05 | 2022-07-08 | 广州汽车集团股份有限公司 | Automatic exposure parameter adjusting method and device, storage medium and shooting equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208293A (en) * | 2014-06-16 | 2015-12-30 | 杭州海康威视数字技术股份有限公司 | Automatic exposure control method of digital camera and device |
| CN105516589A (en) * | 2015-12-07 | 2016-04-20 | 凌云光技术集团有限责任公司 | Intelligent exposure method and system based on face recognition |
| CN106385542A (en) * | 2016-10-11 | 2017-02-08 | 广东欧珀移动通信有限公司 | Camera focusing method, device and mobile terminal |
| US20170171472A1 (en) * | 2015-12-11 | 2017-06-15 | Wistron Corporation | Method and Related Camera Device for Generating Pictures with Object Moving Trace |
| CN107872613A (en) * | 2016-09-23 | 2018-04-03 | 中兴通讯股份有限公司 | A kind of method, device and mobile terminal that recognition of face is carried out using dual camera |
| CN108197601A (en) * | 2018-01-29 | 2018-06-22 | 深圳市巨龙创视科技有限公司 | A kind of Intelligent human-face tracks exposure system |
| CN109068067A (en) * | 2018-08-22 | 2018-12-21 | Oppo广东移动通信有限公司 | Exposal control method, device and electronic equipment |
| CN109889695A (en) * | 2019-02-27 | 2019-06-14 | 努比亚技术有限公司 | A kind of image-region determines method, terminal and computer readable storage medium |
| CN109918993A (en) * | 2019-01-09 | 2019-06-21 | 杭州中威电子股份有限公司 | A kind of control method based on human face region exposure |
| CN110213484A (en) * | 2019-05-31 | 2019-09-06 | 维沃移动通信有限公司 | A kind of photographic method, terminal device and computer readable storage medium |
-
2019
- 2019-11-07 CN CN201911084025.3A patent/CN111131693B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208293A (en) * | 2014-06-16 | 2015-12-30 | 杭州海康威视数字技术股份有限公司 | Automatic exposure control method of digital camera and device |
| CN105516589A (en) * | 2015-12-07 | 2016-04-20 | 凌云光技术集团有限责任公司 | Intelligent exposure method and system based on face recognition |
| US20170171472A1 (en) * | 2015-12-11 | 2017-06-15 | Wistron Corporation | Method and Related Camera Device for Generating Pictures with Object Moving Trace |
| CN107872613A (en) * | 2016-09-23 | 2018-04-03 | 中兴通讯股份有限公司 | A kind of method, device and mobile terminal that recognition of face is carried out using dual camera |
| CN106385542A (en) * | 2016-10-11 | 2017-02-08 | 广东欧珀移动通信有限公司 | Camera focusing method, device and mobile terminal |
| CN108197601A (en) * | 2018-01-29 | 2018-06-22 | 深圳市巨龙创视科技有限公司 | A kind of Intelligent human-face tracks exposure system |
| CN109068067A (en) * | 2018-08-22 | 2018-12-21 | Oppo广东移动通信有限公司 | Exposal control method, device and electronic equipment |
| CN109918993A (en) * | 2019-01-09 | 2019-06-21 | 杭州中威电子股份有限公司 | A kind of control method based on human face region exposure |
| CN109889695A (en) * | 2019-02-27 | 2019-06-14 | 努比亚技术有限公司 | A kind of image-region determines method, terminal and computer readable storage medium |
| CN110213484A (en) * | 2019-05-31 | 2019-09-06 | 维沃移动通信有限公司 | A kind of photographic method, terminal device and computer readable storage medium |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112866773A (en) * | 2020-08-21 | 2021-05-28 | 海信视像科技股份有限公司 | Display device and camera tracking method in multi-person scene |
| CN112866773B (en) * | 2020-08-21 | 2023-09-26 | 海信视像科技股份有限公司 | Display equipment and camera tracking method in multi-person scene |
| CN112511749A (en) * | 2020-11-30 | 2021-03-16 | 上海摩象网络科技有限公司 | Automatic exposure control method and device for target object and electronic equipment |
| CN112839183A (en) * | 2020-12-28 | 2021-05-25 | 杭州宇泛智能科技有限公司 | Environment self-adaptive face image recognition method |
| CN114727024A (en) * | 2021-01-05 | 2022-07-08 | 广州汽车集团股份有限公司 | Automatic exposure parameter adjusting method and device, storage medium and shooting equipment |
| CN112911146A (en) * | 2021-01-27 | 2021-06-04 | 杭州寰宇微视科技有限公司 | Intelligent dimming method based on human face |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111131693B (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111131693B (en) | Face image enhancement method based on multi-exposure face detection | |
| CN105915811B (en) | The strong Xanthophyll cycle adaptive control method of image and device | |
| CN108174118B (en) | Image processing method and device and electronic equipment | |
| CN101448085B (en) | Videography processing method and system supporting face detection | |
| CN101282425B (en) | Method and device for compensating backlight | |
| US20100103194A1 (en) | Method and system for fusing images | |
| CN111028190A (en) | Image processing method, image processing device, storage medium and electronic equipment | |
| CN112752023B (en) | Image adjusting method and device, electronic equipment and storage medium | |
| CN111770285B (en) | Exposure brightness control method and device, electronic equipment and storage medium | |
| CN104580925A (en) | Image brightness controlling method, device and camera | |
| US20100232685A1 (en) | Image processing apparatus and method, learning apparatus and method, and program | |
| US20080043112A1 (en) | Exposure of Digital Imaging | |
| CN106357987A (en) | Exposure method and device | |
| CN110248112A (en) | A kind of exposal control method of imaging sensor | |
| CN109766828A (en) | A kind of vehicle target dividing method, device and communication equipment | |
| CN107241558A (en) | Exposure processing method, device and terminal device | |
| CN109639992B (en) | Projection method based on automatic exposure control and automatic switching of projection patterns | |
| CN105208293A (en) | Automatic exposure control method of digital camera and device | |
| CN111225162B (en) | Image exposure control method, system, readable storage medium and camera equipment | |
| CN110807406B (en) | Foggy day detection method and device | |
| CN112653845A (en) | Exposure control method, exposure control device, electronic equipment and readable storage medium | |
| CN106686320B (en) | A kind of tone mapping method based on number density equilibrium | |
| CN104915933A (en) | Foggy day image enhancing method based on APSO-BP coupling algorithm | |
| CN103618846A (en) | Background removing method for restricting influence of sudden changes of light in video analysis | |
| CN110099207B (en) | Effective image calculation method for overcoming camera instability |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |