CN110889801B - Camera decontamination optimization method of smoke kitchen system and smoke kitchen system - Google Patents
Camera decontamination optimization method of smoke kitchen system and smoke kitchen system Download PDFInfo
- Publication number
- CN110889801B CN110889801B CN201810931753.2A CN201810931753A CN110889801B CN 110889801 B CN110889801 B CN 110889801B CN 201810931753 A CN201810931753 A CN 201810931753A CN 110889801 B CN110889801 B CN 110889801B
- Authority
- CN
- China
- Prior art keywords
- image
- camera
- area
- pollutant
- decontamination
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 239000000779 smoke Substances 0.000 title claims abstract description 34
- 238000005202 decontamination Methods 0.000 title claims abstract description 29
- 230000003588 decontaminative effect Effects 0.000 title claims abstract description 29
- 238000005457 optimization Methods 0.000 title claims abstract description 19
- 239000003344 environmental pollutant Substances 0.000 claims abstract description 54
- 231100000719 pollutant Toxicity 0.000 claims abstract description 54
- 239000000356 contaminant Substances 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 20
- 238000010411 cooking Methods 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000011109 contamination Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 235000019504 cigarettes Nutrition 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 4
- 230000000903 blocking effect Effects 0.000 claims description 2
- 238000002203 pretreatment Methods 0.000 claims 1
- 238000004140 cleaning Methods 0.000 description 12
- 230000002159 abnormal effect Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 239000003595 mist Substances 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- G06T5/73—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration by the use of histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Abstract
The invention discloses a camera decontamination optimization method of a smoke kitchen system, which comprises the following steps: a condition judging step, namely acquiring a current image after the camera is started, acquiring image data of the current image, judging whether the current image meets a preset condition, and if so, entering a pollution degree determining step; a pollution degree determining step, namely comparing the image data with the pre-acquired original data to evaluate the image quality of the current image and determine whether the pollution degree of the camera lens exceeds a threshold value, and if not, entering a pollutant determining step; a pollutant determining step of determining a pollutant position of pollutants on a camera lens; and a decontamination optimizing step, wherein the image of the region corresponding to the dirt position shot by the camera is processed according to the dirt position so as to optimize the image quality. By adopting the method provided by the invention, the application problem of the camera in a smoke kitchen system can be solved, and a user is reminded to clean the camera in time or optimize the image quality through an algorithm.
Description
Technical Field
The invention relates to the field of smoke exhaust stoves, in particular to a camera decontamination optimization method of a smoke exhaust stove system. The invention also discloses a smoke kitchen system.
Background
With the maturation of intelligent technology, more and more household appliances begin to carry intelligent modules to expand self intelligent functions. In the field of kitchen ranges of smoke machines, a typical intelligent module application comprises a camera, and by means of an image recognition technology, not only can the recognition of food materials and cooking states be realized, but also the smoke concentration can be monitored through image information, so that the intelligent degree of a kitchen range product is greatly improved.
However, a large amount of oil smoke is easily generated in a kitchen environment, and the corresponding oil smoke can adhere to the lens of the camera, so that the shooting performance of the camera is affected. Meanwhile, in order to ensure the normal work of the camera, the user needs to be frequently reminded of cleaning, and the cleaning is too frequent to seriously influence the user experience of the user on the smoke kitchen product.
Disclosure of Invention
The invention provides a camera decontamination optimization method for a smoke kitchen system, which can intelligently detect whether a camera lens is polluted or not, remind a user to clean in time according to the pollution degree of the camera lens, or optimize shooting quality through an image algorithm under the condition that the pollution is not serious. The invention also discloses a smoke kitchen system which is provided with the camera and can intelligently detect the dirt degree of the camera lens, thereby reminding a user to clean or automatically optimize shooting quality and improving the applicability of the camera in the smoke kitchen system.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a camera decontamination optimization method of a smoke kitchen system comprises the following steps: a condition judging step, namely acquiring a current image after the camera is started, acquiring image data of the current image, judging whether the current image meets a preset condition, and if so, entering a pollution degree determining step; a pollution degree determining step, namely comparing the image data with the pre-acquired original data to evaluate the image quality of the current image and determine whether the pollution degree of the camera lens exceeds a threshold value, and if not, entering a pollutant determining step; a pollutant determining step of determining a pollutant position of pollutants on a camera lens; and a decontamination optimizing step, wherein the image of the region corresponding to the dirt position shot by the camera is processed according to the dirt position so as to optimize the image quality.
Further, the contaminant determination step includes: performing color space conversion on a current image, and dividing the converted image into M x N sub-area images through blocking; and respectively performing a first processing flow and a second processing flow on the M-N sub-area images.
Further, the first process flow includes: the sub-area images are respectively decomposed into three-channel gray-scale images Yi, ui and Vi, DCT conversion is carried out on the gray-scale images Yi, ui and Vi, threshold correction is carried out on the converted images, DCT inverse transformation is carried out on the corrected images Yi ', ui and Vi', and then an average value PSNR_mean of peak signal-to-noise ratios and an average value SSIM_mean of structural similarity of the corresponding sub-area images i are obtained through calculation, wherein i represents the ith sub-area image.
Further, the contaminant determination step further includes: the type of contamination on the camera lens is determined.
Further, the second process includes: respectively carrying out color histogram calculation on the sub-region images to obtain corresponding sub-region histogram data; carrying out correlation calculation on the sub-region histogram data and preset pollutant histogram data, and respectively obtaining similarity coefficients Ci of the corresponding sub-region image i and the corresponding pollutants; and determining the pollutant type and the pollutant position according to the similarity coefficient Ci and the corresponding subarea image.
Further, the decontamination optimization step includes: determining whether the condition PSNR_mean < Tpsnr, SSIM_mean < Tssim, and Ci > Tc is satisfied at the dirty position according to the type of the pollutant and the dirty position; if yes, recalculating the gray value of the corresponding pixel point of the corresponding region and replacing the image of the region through interpolation processing; if not, reserving the gray value of the corresponding pixel point of the corresponding area and judging that the area is not affected by the pollutant; wherein Tpsnr is a preset peak signal-to-noise ratio threshold, tsm is a preset structural similarity threshold, and Tc is a preset similarity coefficient threshold.
Further, the method further comprises: and initializing, namely acquiring a single frame image as a reference image after the camera is started for the first time, preprocessing the reference image to acquire and record the original data containing the hearth area, wherein the first area of the hearth area in the reference image is S1.
Further, the preset conditions include: at least part of the current image comprises a hearth area, the second area of the hearth area in the current image is S2, S2 is less than or equal to alpha and less than or equal to S1, alpha is more than or equal to 0.5 and less than or equal to 1, and alpha is a preset area shielding coefficient.
Further, the preprocessing includes: positioning treatment of a cooking bench area and matting treatment of the cooking bench area.
The invention also discloses a smoke kitchen system:
the utility model provides a cigarette kitchen system, includes cigarette machine and cooking utensils, is provided with camera, warning module and decontamination optimizing module in the cigarette kitchen system, and the camera is used for gathering the image of culinary art in-process, and decontamination optimizing module is used for handling the image of the dirty position corresponding region of taking with optimizing image quality, reminds the module to be used for sending the unusual warning of camera and/or the clean warning of camera to the user.
The technical scheme of the invention has the following beneficial effects:
the camera decontamination optimization method of the smoke kitchen system disclosed by the invention can identify the type of pollutants and the dirty positions, remind a user to clean under the condition of serious dirt, and replace a dirty area by using a corresponding image algorithm under the condition of not serious dirt, thereby reducing the times of cleaning the camera by the user and improving the shooting quality. The invention also discloses a smoke kitchen system which not only can prompt a user to clean the camera in time, but also can optimize shooting quality through an algorithm, and has higher intelligence compared with the current smoke kitchen system.
Drawings
FIG. 1 is a block diagram of the steps of one embodiment of the method of the present invention;
FIG. 2 is a schematic flow chart of one embodiment of the method of the present invention;
FIG. 3 is a schematic flow chart of a further embodiment of the method of the present invention;
FIG. 4 is a schematic illustration of image replacement in areas in one embodiment of the method of the present invention;
fig. 5 is a block diagram of a range system according to the present invention.
Detailed Description
The technical scheme provided by the invention is described in more detail below through the attached drawings and specific embodiments:
FIG. 1 discloses a block diagram of steps of an embodiment of the method of the present invention, which discloses a camera decontamination optimization method of a range system, comprising:
step 101: a condition judging step, namely acquiring a current image after the camera is started, acquiring image data of the current image, judging whether the current image meets a preset condition, and if so, entering a pollution degree determining step;
step 102: a pollution degree determining step, namely comparing the image data with the pre-acquired original data to evaluate the image quality of the current image and determine whether the pollution degree of the camera lens exceeds a threshold value, and if not, entering a pollutant determining step;
step 103: a pollutant determining step of determining a pollutant position of pollutants on a camera lens;
step 104: and a decontamination optimizing step, wherein the image of the region corresponding to the dirt position shot by the camera is processed according to the dirt position so as to optimize the image quality.
In this embodiment, a camera decontamination optimization method of a range system is disclosed, which includes a condition determination step, a contamination level determination step, a contamination determination step, and a decontamination optimization step. When the smoke kitchen system needs to work, the current image is acquired after the camera is started, and whether the camera is shielded or covered by a large amount of fog or other abnormal conditions is determined through a condition judging step, so that the situation that the camera is in an abnormal working environment is eliminated; and through the dirty degree determining step, whether the user needs to clean the camera can be judged, and the pollutant determining step and the decontamination optimizing step are carried out under the condition that the influence of pollutants is eliminated through an algorithm, and the shooting quality of the image is optimized through the algorithm, so that the frequency of cleaning the camera by the user is reduced.
In one embodiment of the method of the present invention, the condition determining step further includes: if not, sending out abnormal reminding of the camera to the user. In this embodiment, the camera may be blocked in a large area, or a large amount of water mist (for example, when the cover is uncovered in the cooking process) is covered, so that the camera cannot work normally, at this time, the user should be reminded that the camera is in an abnormal state, the corresponding adverse factors cannot be eliminated through the algorithm, and at this time, the camera cannot judge whether the camera is dirty or not, so that the user should be reminded of eliminating the influence of the working environment, so as to restore the normal working state. In the embodiment, the intelligent reminding can be carried out on the user, the abnormality of the camera is distinguished, and the misjudgment of the dirt degree of the camera is avoided.
In one embodiment of the method of the present invention, the contamination level determining step further includes: and if yes, sending a camera cleaning prompt to the user. In this embodiment, if the contamination level of the camera is found to exceed the threshold value through detection, the influence of the contamination on the image capturing is difficult to be compensated by the algorithm, and the user should be reminded to clean the camera in time at this time, so as to avoid influencing the capturing parameters and the accuracy of image recognition. Compared with the prior art, the scheme in the embodiment can autonomously judge the dirt degree of the camera and prompt in time, so that the problem that the dirt degree of the camera cannot be determined by a user in the past is solved.
As shown in fig. 2, a schematic flow chart of one embodiment of the method according to the present invention is involved. The embodiment discloses a specific flow of a camera decontamination optimization method of a smoke kitchen system, which comprises the following steps:
step 201, initializing a camera: acquiring a reference image and recording original data;
step 202, whether the image meets the preset condition: determining whether the preset condition is met according to the area of the hearth area, if so, executing step 203, otherwise, executing step 207;
step 203, determining whether the dirt level exceeds a threshold value: comparing the image data with the original data to determine whether the dirt degree exceeds a threshold value, if so, executing step 208, otherwise, executing step 204;
step 204, determining a contaminant location and/or a contaminant type: determining the type of the pollutant at the corresponding pollutant position through correlation calculation;
step 205, whether PSNR_mean < Tpsnr, SSIM_mean < Tssim, and Ci > Tc are satisfied: judging that the replacement processing is needed, if yes, executing a step 206, and if not, executing a step 209;
step 206, replacing the image of the area corresponding to the dirty position: replacing the image of the corresponding region to optimize image quality;
step 207, sending out an abnormal camera reminder: reminding a user to eliminate interference factors;
step 208, sending a camera cleaning prompt: reminding a user to clean the camera;
step 209, the corresponding area is not affected, and no processing is performed: no replacement of this area is required.
The embodiment discloses a specific flow of the method, and the method in the embodiment can not only remind a user of eliminating environmental interference or timely cleaning the camera according to conditions, but also replace images at the positions corresponding to pollutants through an algorithm under the condition that the pollution degree is not serious, so that the frequency of cleaning the camera by the user can be reduced, and the images shot by the camera are optimized to avoid affecting the accuracy of image identification.
As one embodiment of the method of the present invention, the contaminant determining step includes:
performing color space conversion on the current image, and dividing the converted image into M x N sub-area images through block processing;
and respectively carrying out a first processing flow and a second processing flow on the M x N sub-area images.
In this embodiment, in order to determine the dirty position of the contaminant on the camera lens, the current image needs to be subjected to color space conversion and segmentation, and the dirty position of the corresponding contaminant can be obtained through the subsequent first processing flow and the second processing flow. Preferably, the current image is converted to a YUV color space to further determine the type of contaminant in conjunction with a subsequent algorithm.
As an embodiment of the method of the present invention, the first process flow includes:
decomposing the subarea image into three-channel gray level images Yi, ui and Vi respectively, performing DCT on the gray level images Yi, ui and Vi, performing threshold correction on the converted images, performing DCT inverse transformation on the corrected images Yi ', ui and Vi', and calculating to obtain an average value PSNR_mean of peak signal to noise ratio and an average value SSIM_mean of structural similarity of the corresponding subarea image i, wherein i represents the ith subarea image.
In the embodiment, each segmented sub-area image is decomposed into three-channel gray level images Yi, ui and Vi respectively, DCT transformation is carried out to obtain three DCT transformation images Yi ', ui ' and Vi ' of the corresponding sub-area image i, and threshold correction processing is carried out on the three DCT transformation images of the sub-area image i respectively; performing inverse DCT transformation on the corrected Yi ', ui ', vi ' to obtain three new graphs Yi1, ui1 and Vi1 related to the subarea image i; the PSNR parameter (peak signal to noise ratio) and the SSIM parameter (structural similarity) of Yi1 are calculated respectively, the PSNR parameter and the SSIM parameter of Ui1, the PSNR parameter and the SSIM parameter of Vi1 are averaged to obtain an average value PSNR_mean of the peak signal to noise ratio of the sub-area image i, and the average value SSIM_mean of the peak signal to noise ratio of the sub-area image i is obtained to average the three SSIM parameters corresponding to the sub-area image i for subsequent calculation and judgment.
In one embodiment of the method of the present invention, the contaminant determination step further comprises: determining the type of contamination on the camera lens.
In this embodiment, the location of the contaminant can be roughly determined by comparing the image data with the original data. Furthermore, in this embodiment, the contaminant determining step may determine not only the location of the contaminant on the camera lens, but also the type of contaminant, so that an appropriate algorithm and parameters are selected according to the type of contaminant to replace the image of the corresponding contaminated area.
In one embodiment of the method of the present invention, the second process includes:
respectively carrying out color histogram calculation on the subarea images to obtain corresponding subarea histogram data;
performing correlation calculation on the sub-region histogram data and preset pollutant histogram data, and respectively obtaining similarity coefficients Ci of the corresponding sub-region image i and the corresponding pollutants;
and determining the pollutant type and the pollutant position according to the similarity coefficient Ci and the corresponding subarea image.
In the embodiment, performing color histogram calculation on the sub-region image i to obtain sub-region histogram data of the sub-region image i; carrying out correlation calculation on the sub-region histogram data of the corresponding sub-region image i and preset pollutant histogram data, comparing the similarity of the sub-region histogram data and the corresponding pollutant histogram data, and calculating to obtain a similarity coefficient Ci of the sub-region image i and the corresponding pollutant; the type of the contaminant and the position of the dirt can be determined according to the similarity coefficient Ci and the position of the corresponding sub-area image i. For example, after obtaining the sub-area histogram data of a sub-area image, performing correlation calculation on the sub-area histogram data and pre-measured pollutant histogram data, wherein the pollutants comprise greasy dirt, smoke dust, water mist and the like, respectively calculating and obtaining the similarity between the sub-area histogram data and the pollutant histogram data of each corresponding type, and taking the similarity coefficient Ci with the highest similarity value as the similarity coefficient i of the sub-area image i and the corresponding pollutant, thereby determining the type of the pollutant corresponding to the position of the sub-area image i. For example, by calculation, at the sub-region image i, the oil stain histogram data with the highest similarity is taken as the corresponding similarity coefficient Ci between the sub-region histogram data of the sub-region image and the oil stain histogram data.
In one embodiment of the method of the present invention, the decontamination optimization step comprises:
determining whether the dirty position meets the condition according to the pollutant type and the dirty position
PSNR_mean < Tpsnr, SSIM_mean < Tssim, and Ci > Tc;
if yes, recalculating the gray value of the corresponding pixel point of the corresponding region and replacing the image of the region through interpolation processing; if not, reserving the gray value of the corresponding pixel point of the corresponding area and judging that the area is not affected by the pollutant; wherein Tpsnr is a preset peak signal-to-noise ratio threshold, tsm is a preset structural similarity threshold, and Tc is a preset similarity coefficient threshold.
In this embodiment, the decontamination optimization step further comprises determining whether replacement of the corresponding image region at the corresponding sub-region image i is required based on conditions, with a specific determination rule that
PSNR_mean < Tpsnr, SSIM_mean < Tssim, and Ci > Tc; wherein Tpsnr is a preset peak signal-to-noise ratio threshold, tsm is a preset structural similarity threshold, and Tc is a preset similarity coefficient threshold. In this embodiment, the type of contaminant most likely to exist in the sub-area image i is determined according to Ci, and then the above condition is combined to determine, if the condition is satisfied, it indicates that the area is indeed a dirty location, and the type of contaminant can also be determined by Ci, so that replacement is required. Because the different types of pollutants can cause the change of parameters such as image definition, gray scale and the like, the replacement of the image of the corresponding area at the dirty position can be specifically adjusted according to the type of the corresponding pollutant. In this embodiment, if the above conditions are all met, it is indicated that the contamination is actually present here, and the image at the location needs to be optimized for replacement according to the specific type of contamination; if any of the images cannot be satisfied, the image is not polluted, replacement or optimization is not needed, and the original image is reserved and is not processed.
As shown in fig. 3, a schematic flow chart of a further embodiment of the method according to the present invention is shown. In this embodiment, a method for optimizing decontamination of a camera of a range system is disclosed, comprising:
step 301, collecting an image: the camera acquires a current image and acquires image data;
step 302, cooktop area detection: detecting a hearth area to determine whether a current image meets a preset condition;
step 303, successfully detecting a cooking bench: judging whether the hearth area is successfully detected, if so, entering a step 304, and if not, entering a step 309;
step 304, image quality evaluation: evaluating the image quality of the current image;
step 305, comparing with the original data: comparing the image data with the original data;
step 306, exceeding the threshold: judging whether the dirt degree exceeds a threshold value, if so, entering a step 310, otherwise, entering a step 307;
step 307, contaminant type determination: judging the type of the pollutant, for example, the pollutant is greasy dirt;
step 308, the algorithm eliminates the greasy dirt area: removing the region polluted by the greasy dirt through an algorithm and replacing the region;
step 309, abnormal camera start prompt: sending an abnormal camera prompt to a user;
step 310, camera oil stain is out of range, and starting prompt: and sending a camera cleaning prompt to a user.
In this embodiment, steps 304 to 306 involve a contamination level determining step, in which image data of the current image and original data of the reference image can be compared by preset image parameters, and whether the corresponding image parameter variation level of the current image exceeds a threshold value compared with the reference image is checked, so as to represent the contamination level of the camera lens, and initially evaluate the image quality of the current image. Wherein the image parameters include, but are not limited to, sharpness, contrast, peak signal-to-noise ratio, and/or structural similarity of the image, etc.
In this embodiment, step 308 involves a decontamination optimization step, wherein FIG. 4 is a schematic diagram of a region image replacement in accordance with one embodiment of the method of the present invention. As shown in fig. 4, after determining that the areas where the dirt positions A0 to A8 are located and the type of the contaminants is greasy dirt, the areas where A0 to A8 are located are replaced. In this embodiment, a linear interpolation replacement method is adopted to replace an image of a corresponding area at a dirty position, specifically, gray values of nine pixel points A0 to A8 need to be recalculated, where:
A0＝α 0 R1+β 0 R8，A1＝α 1 R2+β 1 R7，
A2＝α 2 R3+β 2 R6，A3＝α 3 R4+β 3 R5，
A4＝α 4 R2+β 4 R7+λR4+μR5，
A5＝α 5 R4+β 5 R5，A6＝α 6 R3+β 6 R6，
A7＝α 7 R2+β 7 R7，A8＝α 8 R1+β 8 R8；
wherein Ri is the gray average value of the corresponding block, α i 、β i Lambda, mu are linear coefficients related to the relative positions of pixel points and the gray value change rate, and the dirt position is recalculated by the above methodThe gray value of the corresponding pixel point of the corresponding region is positioned, so that the image of the corresponding region can be replaced under the condition that the pollution is not serious enough, and the influence of pollutants is eliminated. For the uncontaminated block, only the gray value of the corresponding pixel point of the corresponding region is reserved, and the replacement processing is not needed.
In one embodiment of the method of the present invention, the method further comprises:
in the initialization step, a single frame image is collected as a reference image after the camera is started for the first time, the reference image is preprocessed to obtain and record original data containing a cooking bench area, and the first area of the cooking bench area in the reference image is S1.
The method in this embodiment also involves an initialization step aimed at acquiring a reference image, detecting the hob area and recording the raw data for subsequent condition decisions and soil level determinations. The reference image is acquired when the camera is started for the first time, and is usually used for the first startup of a newly purchased smoke stove system, and the camera is not polluted, so that the reference image can contain a complete hearth area, and corresponding original data are accurate, and can be used for subsequent condition judgment and inspection. In this embodiment, the first area of the cooktop area in the reference image is S1.
In one embodiment of the method of the present invention, the preset condition includes:
the current image at least partially comprises the hearth area, the second area of the hearth area in the current image is S2, S2 is less than or equal to alpha and equal to S1, alpha is more than or equal to 0.5 and less than or equal to 1, and alpha is a preset area shielding coefficient.
The embodiment relates to the condition determining step of determining whether or not the current image satisfies the preset condition based on the second area of the hearth area included in the current image and the first area of the hearth area included in the reference image, since the reference image has already been acquired by the smoke kitchen system, when the smoke kitchen system is turned on again by the user. The reason that s2+_αs1 is required is that the complete range area can be theoretically photographed when the camera is first turned on, and if the camera is turned on again, the camera cannot work normally if the camera is blocked or suddenly forms a large amount of water mist on the lens, and the area of the range area photographed by the camera is necessarily smaller than S1. If only a small portion of the area of the hob area is not detected, the subsequent steps can be continued, but if a large portion of the area of the hob area is not detected, it is indicated that the camera is in an abnormal working state, and the user needs to be reminded of eliminating the influence of the interference factors, so in this embodiment, whether the camera is in an abnormal working state is determined through the first area S1 and the second area S2, wherein α is a preset area shielding coefficient, and α is preferably set in a range of 0.5.ltoreq.α.ltoreq.1.
In one embodiment of the method of the present invention, the preprocessing includes: and positioning the hearth area and matting the hearth area. The embodiment relates to an initialization step, wherein the preprocessing comprises positioning processing of a cooking bench area and matting processing of the cooking bench area, namely the position of the cooking bench area in a reference image needs to be determined, and the outline of the cooking bench area is determined for matting, so that subsequent comparison is facilitated.
As shown in fig. 5, a module architecture diagram relating to a range system according to the present invention. The invention also discloses a smoke kitchen system 500 which comprises a smoke machine and a kitchen range, wherein the smoke kitchen system is provided with a camera 501, a reminding module 502 and a decontamination optimizing module 503, the camera 501 is used for collecting images in the cooking process, the decontamination optimizing module 503 is used for processing the images of the corresponding areas of the dirt positions shot by the camera to optimize the image quality, and the reminding module 502 is used for sending abnormal reminding and/or cleaning reminding of the camera to a user. In this embodiment, the camera 502 is preferably arranged on the range, the range area is photographed downwards, the range system therein is able to recognize the cooking food material by means of the camera 502, and the smoke concentration generated by cooking is determined from the photographed image. The smoke kitchen system 500 in this embodiment can also send out a camera abnormality alert when the camera 502 is in an abnormal operating state, and can also send out a camera cleaning alert to the user when the camera 502 needs manual cleaning.
The above embodiments are only for illustrating the design method of the present invention, and are not intended to limit the protection scope of the present invention. Modifications and transformations under the teaching of the present invention should be considered as falling within the scope of the present invention.
Claims (8)
1. The method for optimizing the decontamination of the camera of the smoke kitchen system is characterized by comprising the following steps of:
a condition judging step, namely acquiring a current image after the camera is started, acquiring image data of the current image, judging whether the current image meets a preset condition, and if yes, entering a pollution degree determining step;
a pollution degree determining step, namely comparing the image data with pre-acquired original data to evaluate the image quality of the current image and determine whether the pollution degree of the camera lens exceeds a threshold value, and if not, entering a pollutant determining step;
a pollutant determining step, namely performing color space conversion on the current image, dividing the converted image into M x N sub-area images through blocking processing, and respectively performing a first processing flow and a second processing flow on the M x N sub-area images to determine the pollutant positions of pollutants on the camera lens;
the first process flow comprises: decomposing the subarea image into three-channel gray level images Yi, ui and Vi respectively, performing DCT on the gray level images Yi, ui and Vi, performing threshold correction on the converted images, performing DCT inverse transformation on the corrected images Yi ', ui ' and Vi ', and calculating to obtain an average value PSNR_mean of peak signal to noise ratio and an average value SSIM_mean of structural similarity of the corresponding subarea image i, wherein i represents an ith subarea image;
the second treatment process is used for determining the pollutant type and the pollutant position;
and a decontamination optimizing step, wherein the image of the area corresponding to the dirt position shot by the camera is processed according to the dirt position so as to optimize the image quality.
2. The method of claim 1, wherein the contaminant determination step further comprises: determining the type of contamination on the camera lens.
3. The method of claim 2, wherein the second process flow comprises:
respectively carrying out color histogram calculation on the subarea images to obtain corresponding subarea histogram data;
performing correlation calculation on the sub-region histogram data and preset pollutant histogram data, and respectively obtaining similarity coefficients Ci of the corresponding sub-region image i and the corresponding pollutants;
and determining the pollutant type and the pollutant position according to the similarity coefficient Ci and the corresponding subarea image.
4. A method according to claim 3, wherein the decontamination optimization step comprises:
determining whether a condition PSNR_mean < Tpsnr, SSIM_mean < Tssim, and Ci > Tc is satisfied at a dirty position according to the type of the contaminant and the dirty position;
if yes, recalculating the gray value of the corresponding pixel point of the corresponding region and replacing the image of the region through interpolation processing; if not, reserving the gray value of the corresponding pixel point of the corresponding area and judging that the area is not affected by the pollutant; wherein Tpsnr is a preset peak signal-to-noise ratio threshold, tsm is a preset structural similarity threshold, and Tc is a preset similarity coefficient threshold.
5. The method according to claim 1, wherein the method further comprises:
and initializing, namely acquiring a single frame image as a reference image after the camera is started for the first time, preprocessing the reference image to acquire and record original data containing a cooking bench area, wherein the first area of the cooking bench area in the reference image is S1, and judging whether the current image meets the preset condition or not.
6. The method of claim 5, wherein the preset conditions include:
the current image at least partially comprises the hearth area, the second area of the hearth area in the current image is S2, S2 is less than or equal to alpha and equal to S1, alpha is more than or equal to 0.5 and less than or equal to 1, and alpha is a preset area shielding coefficient; and when S2 is less than or equal to alpha and S1, the current image meets the preset condition.
7. The method according to claim 5 or 6, wherein the pre-treatment comprises: and positioning the hearth area and matting the hearth area.
8. A range hood system comprising a range hood and a range, for performing the method of any of claims 1-7; be provided with camera, warning module and decontamination optimization module in the cigarette kitchen system, the camera is used for gathering the image of culinary art in-process, decontamination optimization module is used for handling the image of the dirty position corresponding region of camera shooting in order to optimize image quality, the warning module is used for sending the unusual warning of camera and/or the clean warning of camera to the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810931753.2A CN110889801B (en) | 2018-08-16 | 2018-08-16 | Camera decontamination optimization method of smoke kitchen system and smoke kitchen system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810931753.2A CN110889801B (en) | 2018-08-16 | 2018-08-16 | Camera decontamination optimization method of smoke kitchen system and smoke kitchen system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110889801A CN110889801A (en) | 2020-03-17 |
| CN110889801B true CN110889801B (en) | 2023-10-20 |
Family
ID=69744133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810931753.2A Active CN110889801B (en) | 2018-08-16 | 2018-08-16 | Camera decontamination optimization method of smoke kitchen system and smoke kitchen system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110889801B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111182294B (en) * | 2020-01-06 | 2021-11-30 | 杭州九阳小家电有限公司 | Intelligent household appliance control method for improving image quality and intelligent household appliance |
| CN111524125B (en) * | 2020-04-28 | 2023-09-01 | 京东科技信息技术有限公司 | Equipment cleaning method, device, system, electronic equipment and storage medium |
| CN111882540A (en) * | 2020-07-28 | 2020-11-03 | 歌尔科技有限公司 | Method, device and equipment for detecting stains on camera protective cover |
| CN112215794B (en) * | 2020-09-01 | 2022-09-20 | 北京中科慧眼科技有限公司 | Method and device for detecting dirt of binocular ADAS camera |
| US11368672B2 (en) * | 2020-09-14 | 2022-06-21 | Argo AI, LLC | Validation of a camera cleaning system |
| CN112188190A (en) * | 2020-09-30 | 2021-01-05 | 广东美的厨房电器制造有限公司 | Stain detection method, cooking appliance, server, and storage medium |
| CN112333441A (en) * | 2020-10-28 | 2021-02-05 | 维沃移动通信有限公司 | Camera detection method and device and electronic equipment |
| CN114273315B (en) * | 2022-03-04 | 2022-06-14 | 华电电力科学研究院有限公司 | Demister blockage pre-control system, method, equipment and medium |
| CN115299767A (en) * | 2022-08-04 | 2022-11-08 | 杭州老板电器股份有限公司 | Steam oven and method for preventing water hanging of protective glass |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104539939A (en) * | 2014-12-17 | 2015-04-22 | 惠州Tcl移动通信有限公司 | Lens cleanliness detection method and system based on mobile terminal |
| CN104811690A (en) * | 2015-04-01 | 2015-07-29 | 广东欧珀移动通信有限公司 | Message prompting method and device |
| CN105163110A (en) * | 2015-09-02 | 2015-12-16 | 厦门美图之家科技有限公司 | Camera cleanliness detection method and system and shooting terminal |
| CN105491372A (en) * | 2015-11-24 | 2016-04-13 | 努比亚技术有限公司 | Mobile terminal and information processing method |
| CN105959528A (en) * | 2016-07-09 | 2016-09-21 | 深圳市辰卓科技有限公司 | Working scene switching method and apparatus and test device |
| CN106231297A (en) * | 2016-08-29 | 2016-12-14 | 深圳天珑无线科技有限公司 | The detection method of photographic head and device |
| CN106385579A (en) * | 2016-09-12 | 2017-02-08 | 努比亚技术有限公司 | Camera detection device, method and multi-camera terminal |
| CN106415598A (en) * | 2014-05-27 | 2017-02-15 | 罗伯特·博世有限公司 | Detection, identification, and mitigation of lens contamination for vehicle mounted camera systems |
| CN107743224A (en) * | 2017-09-30 | 2018-02-27 | 奇酷互联网络科技（深圳）有限公司 | The dirty based reminding method of camera lens, system, readable storage medium storing program for executing and mobile terminal |
| CN107749918A (en) * | 2017-09-14 | 2018-03-02 | 深圳天珑无线科技有限公司 | User is prompted to wipe method, mobile terminal and the storage medium of camera lens |
| CN107945158A (en) * | 2017-11-15 | 2018-04-20 | 上海摩软通讯技术有限公司 | A kind of dirty method and device of detector lens |
| CN107975851A (en) * | 2018-01-02 | 2018-05-01 | 广东美的厨房电器制造有限公司 | Range hood and its cleaning control method |
-
2018
- 2018-08-16 CN CN201810931753.2A patent/CN110889801B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106415598A (en) * | 2014-05-27 | 2017-02-15 | 罗伯特·博世有限公司 | Detection, identification, and mitigation of lens contamination for vehicle mounted camera systems |
| CN104539939A (en) * | 2014-12-17 | 2015-04-22 | 惠州Tcl移动通信有限公司 | Lens cleanliness detection method and system based on mobile terminal |
| CN104811690A (en) * | 2015-04-01 | 2015-07-29 | 广东欧珀移动通信有限公司 | Message prompting method and device |
| CN105163110A (en) * | 2015-09-02 | 2015-12-16 | 厦门美图之家科技有限公司 | Camera cleanliness detection method and system and shooting terminal |
| CN105491372A (en) * | 2015-11-24 | 2016-04-13 | 努比亚技术有限公司 | Mobile terminal and information processing method |
| CN105959528A (en) * | 2016-07-09 | 2016-09-21 | 深圳市辰卓科技有限公司 | Working scene switching method and apparatus and test device |
| CN106231297A (en) * | 2016-08-29 | 2016-12-14 | 深圳天珑无线科技有限公司 | The detection method of photographic head and device |
| CN106385579A (en) * | 2016-09-12 | 2017-02-08 | 努比亚技术有限公司 | Camera detection device, method and multi-camera terminal |
| CN107749918A (en) * | 2017-09-14 | 2018-03-02 | 深圳天珑无线科技有限公司 | User is prompted to wipe method, mobile terminal and the storage medium of camera lens |
| CN107743224A (en) * | 2017-09-30 | 2018-02-27 | 奇酷互联网络科技（深圳）有限公司 | The dirty based reminding method of camera lens, system, readable storage medium storing program for executing and mobile terminal |
| CN107945158A (en) * | 2017-11-15 | 2018-04-20 | 上海摩软通讯技术有限公司 | A kind of dirty method and device of detector lens |
| CN107975851A (en) * | 2018-01-02 | 2018-05-01 | 广东美的厨房电器制造有限公司 | Range hood and its cleaning control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110889801A (en) | 2020-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110889801B (en) | Camera decontamination optimization method of smoke kitchen system and smoke kitchen system | |
| CN107975851B (en) | Range hood and cleaning control method thereof | |
| CN107485929B (en) | filter screen pollution detection method and device | |
| JPWO2020031984A1 (en) | Parts inspection method and inspection system | |
| CN109523527B (en) | Method and device for detecting dirty area in image and electronic equipment | |
| CN102547365B (en) | Black edge detection method and device for video image | |
| CN105915840B (en) | A method of the factory smoke discharge based on vision signal monitors automatically | |
| KR102606931B1 (en) | Static soiling detection and correction | |
| CN111064870A (en) | Lens cleaning method, control device and storage medium of visual robot | |
| CN111612773B (en) | Thermal infrared imager and real-time automatic blind pixel detection processing method | |
| JP6139002B2 (en) | How to identify anomalies in images | |
| CN116363126B (en) | Welding quality detection method for data line USB plug | |
| CN113724167A (en) | Self-adaptive acquisition and image processing method for high-definition video information | |
| CN112748116A (en) | Medical gauze surface defect online detection method and device | |
| CN116167969A (en) | Lens smudge detection method, device, vehicle, storage medium and program product | |
| JP4903396B2 (en) | Statistical quality control method and system for print engines | |
| Peris-Fajarnés et al. | Rank-ordered differences statistic based switching vector filter | |
| CN109142176B (en) | Smoke subarea space rechecking method based on space association | |
| JP2003189294A (en) | Image monitoring device | |
| CN110529896B (en) | Correction method for smoke detection of range hood and range hood | |
| CN116509262A (en) | Cleaning equipment and cleaning method thereof | |
| CN113409345A (en) | Method for identifying dirt of lens, processor and household appliance | |
| CN111626339A (en) | Method for detecting abnormal die cavity of injection molding machine with light shadow and jitter influence resistance | |
| CN114663429B (en) | Lens pollutant detection method and device, computer equipment and storage medium | |
| TW200926764A (en) | Video noise reduction method suitable for dark environment |
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 |