CN116405661B - Smart television development performance testing method and device - Google Patents

Smart television development performance testing method and device Download PDF

Info

Publication number
CN116405661B
CN116405661B CN202310474021.6A CN202310474021A CN116405661B CN 116405661 B CN116405661 B CN 116405661B CN 202310474021 A CN202310474021 A CN 202310474021A CN 116405661 B CN116405661 B CN 116405661B
Authority
CN
China
Prior art keywords
screen
extinguishing
audio
test
picture
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
Application number
CN202310474021.6A
Other languages
Chinese (zh)
Other versions
CN116405661A (en
Inventor
杜锋良
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Knott Software Shenzhen Co ltd
Original Assignee
Knott Software Shenzhen Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Knott Software Shenzhen Co ltd filed Critical Knott Software Shenzhen Co ltd
Priority to CN202310474021.6A priority Critical patent/CN116405661B/en
Publication of CN116405661A publication Critical patent/CN116405661A/en
Application granted granted Critical
Publication of CN116405661B publication Critical patent/CN116405661B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/02—Diagnosis, testing or measuring for television systems or their details for colour television signals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • 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
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
    • G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
    • G10L25/60—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for measuring the quality of voice signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/04—Diagnosis, testing or measuring for television systems or their details for receivers
    • 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
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30121—CRT, LCD or plasma display
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/02—Recognising information on displays, dials, clocks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • General Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Biomedical Technology (AREA)
  • Acoustics & Sound (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • Human Computer Interaction (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

The invention relates to the technical field of performance test, and discloses a method for testing development performance of an intelligent television, which comprises the following steps: performing color detection on the intelligent television to obtain a color parameter set, splitting a primary test chart set into a screen-off test chart set and a screen-on test chart set, performing picture feature recognition on the screen-off test chart set to obtain a screen-off defect set, performing region feature recognition on the screen-on test chart set to obtain a screen-on defect set, and updating the screen-on defect set according to the screen-off defect set to obtain a standard screen defect set; performing sound wave reconstruction and audio characteristic recognition on the primary test audio to obtain a distorted pronunciation defect set; and performing color gamut labeling on the color parameter set to obtain a color detection result, and collecting the color detection result, the distorted pronunciation defect set and the standard screen defect set into a performance test result. The invention further provides a device for testing the development performance of the intelligent television. The invention can improve the accuracy of the intelligent television during performance test.

Description

Smart television development performance testing method and device
Technical Field
The invention relates to the technical field of performance testing, in particular to a method and a device for testing development performance of an intelligent television.
Background
The intelligent television is a television product based on network application technology, has an open operating system and a chip, has an open application platform, can realize a bidirectional man-machine interaction function, integrates multiple functions such as video, audio, entertainment and data, and the like, so as to meet the diversified and personalized requirements of users, and along with the improvement of the living standard of people, more families begin to use the intelligent television, but the intelligent television needs to be subjected to performance test before leaving a factory.
The existing television performance test method is mostly based on an artificial performance test method, and defects such as dead spots and popping sounds of a screen are checked by using an artificial to perform video and audio test on the intelligent television, so that the performance test is realized.
Disclosure of Invention
The invention provides a method and a device for testing development performance of an intelligent television, and mainly aims to solve the problem of lower accuracy in performance test of the intelligent television.
In order to achieve the above object, the present invention provides a method for testing development performance of an intelligent television, comprising:
shooting a test signal of the intelligent television to be detected by using a pre-corrected camera to obtain a primary test chart set, performing color detection on the intelligent television by using a preset color analyzer to obtain a color parameter set, and recording the test signal of the intelligent television by using a pre-corrected recorder to obtain primary test audio;
splitting the primary test chart set into a screen-extinguishing test chart set and a bright screen test chart set, and sequentially carrying out image denoising, edge detection, image segmentation and image enhancement on each screen-extinguishing test image in the screen-extinguishing test chart set to obtain a secondary screen-extinguishing chart set;
performing picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set, performing feature extraction on secondary screen-extinguishing graph block sets in the secondary screen-extinguishing graph block set one by one to obtain screen-extinguishing feature sets, performing picture feature recognition on the screen-extinguishing feature sets to obtain screen-extinguishing defect sets, performing region feature recognition on the bright screen test graph set to obtain bright screen defect sets, and updating the bright screen defect sets according to the screen-extinguishing defect sets to obtain standard screen defect sets, wherein the performing picture feature recognition on the screen-extinguishing feature sets to obtain the screen-extinguishing defect sets comprises: selecting standard screen-extinguishing characteristics in the screen-extinguishing characteristic group one by one as target standard screen-extinguishing characteristics, and calculating primary screen-extinguishing vectors corresponding to the target standard screen-extinguishing characteristics by using the following screen-extinguishing vector algorithm:
Wherein Z is the primary screen-extinguishing vector, K is the number of feature vectors of the target standard screen-extinguishing feature, K is the kth, x k Feature vectors representing the kth dimension in the target standard screen feature, alpha, beta and gamma are preset weight matrixes of the target standard screen feature,refers to x k Dimension of beta vector,/->Is the normalized coefficient of the screen-off vector algorithm, T is the transposed function; vector transcoding is carried out on the primary screen-off vectors to obtain primary screen-off semantics, all primary screen-off semantics are collected into a screen-off semantic set, and defect semantics are screened out from the screen-off semantic set to be collected into a screen-off defect set;
sequentially performing bandpass filtering, amplitude adjustment and power amplification operation on the primary test audio to obtain secondary test audio, splitting the secondary test audio into a secondary audio sequence by utilizing a preset audio window, performing sonic reconstruction on the secondary audio sequence to obtain standard audio characteristics, and performing audio characteristic recognition on the standard audio characteristics to obtain a distortion pronunciation defect set;
and performing color gamut labeling on the color parameter set to obtain a color detection result, performing gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result, and collecting the color detection result, the pronunciation detection result and the standard screen defect set into a performance test result.
Optionally, the performing color detection on the smart tv by using a preset color analyzer to obtain a color parameter set includes:
performing global brightness detection on the intelligent television by using a preset color analyzer to obtain a regional brightness set;
extracting peak brightness and average brightness from the regional brightness set, calculating a brightness variance corresponding to the regional brightness set according to the average brightness, and taking the brightness variance as brightness uniformity;
performing global contrast detection on the intelligent television by using the color analyzer to obtain a region contrast set, and taking the average value of the region contrast set as a standard contrast;
performing multi-angle chromaticity detection on the intelligent television by using the color analyzer to obtain a chromaticity parameter set;
and collecting the peak brightness, the average brightness, the brightness uniformity, the standard contrast and the chromaticity parameter set into a chromaticity parameter set.
Optionally, the splitting the primary test atlas into a rest screen test atlas and a bright screen test atlas includes:
selecting primary test pictures in the primary test picture set one by one as target primary pictures, and carrying out graying operation on the target primary pictures to obtain target gray-scale pictures;
Calculating an average value of gray values of all pixels in the target gray picture, and taking the average value as a gray average value of the target gray picture;
judging whether the gray average value is larger than a preset gray threshold value or not;
if yes, adding the target primary picture corresponding to the gray average value into a preset bright screen test chart set;
if not, adding the target primary picture corresponding to the gray average value into a preset screen-extinguishing test chart set.
Optionally, the performing operations of picture denoising, edge detection, picture segmentation and picture enhancement on each screen-extinguishing test picture in the screen-extinguishing test picture set sequentially to obtain a secondary screen-extinguishing picture set includes:
selecting the screen-extinguishing test pictures in the screen-extinguishing test picture set one by one as target screen-extinguishing pictures, and performing median filtering operation on the target screen-extinguishing pictures to obtain denoising screen-extinguishing pictures;
performing edge detection on the denoising and screen-extinguishing picture by using a preset weight-cutting fitting algorithm to obtain a screen fitting edge;
carrying out picture segmentation on the denoising screen-extinguishing picture according to the screen fitting edge to obtain a segmented screen-extinguishing picture;
and calculating a screen-extinguishing gray level histogram of the segmented screen-extinguishing picture, carrying out picture enhancement on the segmented screen-extinguishing picture by using the screen-extinguishing gray level histogram to obtain a secondary screen-extinguishing picture, and collecting all the secondary screen-extinguishing pictures into a secondary screen-extinguishing picture set.
Optionally, the performing edge detection on the denoising and screen-extinguishing picture by using a preset weight-cutting fitting algorithm to obtain a screen fitting edge includes:
extracting a primary screen edge set from the denoising screen picture by using a preset edge erosion algorithm;
fitting each primary screen edge in the primary screen edge set by using the following weight-loss fitting algorithm to obtain a standard fitting edge set:
wherein epsilon refers to a fitting error, R refers to the R-th pixel point corresponding to the primary screen edge, R refers to the total number of the pixel points corresponding to the primary screen edge, and w r Refer to the clipping weight of the r pixel point, ρ, σ is the slope coefficient of the fitted primary screen edge, γ is the displacement coefficient of the fitted primary screen edge, a r Refers to the abscissa, b, of the (r) th pixel point corresponding to the edge of the primary screen r The ordinate of the r pixel point corresponding to the primary screen edge is referred, and lambda refers to the fitting coefficient of the weight-cut fitting algorithm;
and splicing all standard fitting edges in the standard fitting edge set into a screen fitting edge.
Optionally, the performing a picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block group set includes:
Sequentially performing inclination correction and picture lifting operation on each secondary screen-extinguishing picture in the secondary screen-extinguishing picture set to obtain a standard screen-extinguishing picture set;
selecting standard screen-extinguishing pictures in the standard screen-extinguishing picture set one by one as target standard screen-extinguishing pictures, sequentially performing quadtree block, trigeminal tree block and binary tree block operation on the target standard screen-extinguishing pictures to obtain standard screen-extinguishing picture block groups, and collecting all the standard screen-extinguishing picture block groups into a standard screen-extinguishing picture block group set;
selecting standard screen-extinguishing pattern blocks in the standard screen-extinguishing pattern block group one by one as a target standard screen-extinguishing pattern block group, and selecting standard screen-extinguishing pattern blocks in the target standard screen-extinguishing pattern block group one by one as target standard screen-extinguishing pattern blocks;
adding the target standard screen-extinguishing block diagram block into a preset primary screen-extinguishing block diagram block, adding standard screen-extinguishing block diagram blocks corresponding to the target standard screen-extinguishing block diagram block position in each standard screen-extinguishing block diagram block except the target standard screen-extinguishing block diagram block in the standard screen-extinguishing block diagram block set into the primary screen-extinguishing block diagram block to obtain a secondary screen-extinguishing block diagram block, and collecting all secondary screen-extinguishing block diagram blocks into a secondary screen-extinguishing block diagram block set.
Optionally, the performing area feature recognition on the bright screen test atlas to obtain a bright screen defect set includes:
sequentially carrying out image denoising, edge detection, image segmentation and image enhancement on each bright screen test image in the bright screen test image set to obtain a secondary bright screen image set;
performing picture blocking operation on the secondary bright screen image set to obtain a secondary bright screen image block set;
and extracting the characteristics of each secondary bright screen block group in the secondary bright screen block group one by one to obtain a bright screen characteristic group, and identifying the picture characteristics of the bright screen characteristic group to obtain a bright screen defect set.
Optionally, the performing bandpass filtering, amplitude adjustment and power amplification on the primary test audio sequentially to obtain a secondary test audio includes:
carrying out band-pass filtering on the primary test audio by using a weighted recursive average filtering method to obtain primary filtered audio;
sequentially carrying out amplitude limiting filtering and jitter elimination filtering on the primary filtering audio to obtain secondary filtering audio;
and performing audio power amplification operation on the secondary filtering audio to obtain secondary test audio.
Optionally, the performing acoustic wave reconstruction on the secondary audio sequence to obtain standard audio features includes:
Selecting secondary audio in the secondary audio sequence one by one as target secondary audio, and performing audio decomposition on the target secondary audio to obtain a target audio decomposition tree;
and performing audio reconstruction on the target audio decomposition tree by using the following parity reconstruction formula to obtain a target audio reconstruction tree:
wherein (1)>Means +.f in the target audio reconstruction tree>The corresponding audio signal value when the abscissa of the mth tree node in the layer is i, e being the filter coefficient, ">Is a low-pass filter function, ">Is a high pass filter function, ">Means the corresponding audio signal value when the abscissa of the 2m tree node in the j-th layer in the target audio decomposition tree is i, < ->Refers to the +.sup.th in the j-th layer in the target audio decomposition tree>Corresponding audio signal values when the abscissa of each tree node is i;
and extracting sample entropy features of the target audio reconstruction tree to obtain reconstructed audio features, and collecting all the reconstructed audio features into standard audio features.
In order to solve the above problems, the present invention further provides a smart tv development performance testing apparatus, the apparatus comprising:
the data acquisition module is used for shooting test signals of the intelligent television to be detected by using a pre-corrected camera to obtain a primary test chart set, performing color detection on the intelligent television by using a preset color analyzer to obtain a color parameter set, and performing test signal recording on the intelligent television by using a pre-corrected recorder to obtain primary test audio;
The picture enhancement module is used for splitting the primary test image set into a screen-extinguishing test image set and a bright screen test image set, and sequentially carrying out picture denoising, edge detection, picture segmentation and picture enhancement on each screen-extinguishing test picture in the screen-extinguishing test image set to obtain a secondary screen-extinguishing image set;
the screen test module is used for carrying out picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set, carrying out feature extraction on secondary screen-extinguishing graph block sets in the secondary screen-extinguishing graph block set one by one to obtain screen-extinguishing feature sets, carrying out picture feature recognition on the screen-extinguishing feature sets to obtain screen-extinguishing defect sets, carrying out region feature recognition on the bright screen test graph set to obtain bright screen defect sets, and updating the bright screen defect sets according to the screen-extinguishing defect sets to obtain standard screen defect sets, wherein the step of carrying out picture feature recognition on the screen-extinguishing feature sets to obtain the screen-extinguishing defect sets comprises the following steps: selecting standard screen-extinguishing characteristics in the screen-extinguishing characteristic group one by one as target standard screen-extinguishing characteristics, and calculating primary screen-extinguishing vectors corresponding to the target standard screen-extinguishing characteristics by using the following screen-extinguishing vector algorithm:
Wherein Z is the primary screen-extinguishing vector, K is the number of feature vectors of the target standard screen-extinguishing feature, K is the kth, x k Feature vectors representing the kth dimension in the target standard screen feature, alpha, beta and gamma are preset weight matrixes of the target standard screen feature,refers to x k Dimension of beta vector,/->Is the normalized coefficient of the screen-off vector algorithm, T is the transposed function; vector transcoding is carried out on the primary screen-off vectors to obtain primary screen-off semantics, all primary screen-off semantics are collected into a screen-off semantic set, and defect semantics are screened out from the screen-off semantic set to be collected into a screen-off defect set;
the audio test module is used for sequentially carrying out band-pass filtering, amplitude adjustment and power amplification operation on the primary test audio to obtain secondary test audio, splitting the secondary test audio into secondary audio sequences by utilizing a preset audio window, carrying out sonic reconstruction on the secondary audio sequences to obtain standard audio characteristics, and carrying out audio characteristic recognition on the standard audio characteristics to obtain a distortion pronunciation defect set;
the result generation module is used for carrying out color gamut labeling on the color parameter set to obtain a color detection result, carrying out gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result, and collecting the color detection result, the pronunciation detection result and the standard screen defect set into an integration performance test result.
According to the embodiment of the invention, the primary test chart set is obtained by shooting test signals of the intelligent television to be detected by utilizing the pre-corrected camera, so that the picture quality of the primary test chart set can be ensured, the defect recognition precision of a subsequent screen can be improved, the primary color display performance of the intelligent television can be measured by utilizing the pre-corrected camera to obtain the color parameter set by carrying out color detection on the intelligent television, the primary test audio is obtained by utilizing the pre-corrected recorder to carry out test signal recording on the intelligent television, the background noise in the audio of the intelligent television can be removed, the recognition precision of the subsequent distortion pronunciation defects can be improved, the defect problems of the intelligent television in the process of the screen-extinguishing and the screen-lighting can be respectively detected by splitting the primary test chart set into the screen-extinguishing test chart set and the screen-lighting test chart set, the picture denoising, the edge detection, the picture segmentation and the picture enhancement operation can be carried out on each screen-extinguishing test chart in sequence, the defect on the intelligent television can be more obvious, the subsequent characteristic extraction and the characteristic extraction of the secondary picture can be carried out, the defect recognition precision can be further improved, the defect recognition precision can be carried out on the screen-extinguishing test chart set can be carried out, the screen-extinguishing defect can be further, the defect recognition precision can be further improved, and the defect can be respectively recognized by carrying out the screen-extinguishing test conditions can be respectively, and the defect detection can be more convenient, and the defect can be more convenient to be detected.
The secondary test audio is obtained by sequentially carrying out band-pass filtering, amplitude adjustment and power amplification operation on the primary test audio, so that the audio details can be further improved, the accuracy of the subsequent audio feature recognition is improved, the standard audio features are obtained by carrying out sonic reconstruction on the secondary audio sequence, the standard audio features are subjected to audio feature recognition to obtain a distorted pronunciation defect set, the high-frequency detail features and the low-frequency detail features of the reserved audio can be improved, the recognition accuracy of true pronunciation defects is improved, and the intelligent television can be subjected to performance test on three sides of screen colors, screen defects and audio defects by collecting the color detection results, the pronunciation detection results and the standard screen defect set into performance test results, so that the accuracy of performance test is improved. Therefore, the intelligent television development performance testing method and device provided by the invention can solve the problem of lower accuracy in performance testing of the intelligent television.
Drawings
Fig. 1 is a flow chart of a method for testing development performance of an intelligent television according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a process for generating a secondary screen-extinguishing atlas according to an embodiment of the present invention;
Fig. 3 is a schematic flow chart of picture blocking according to an embodiment of the present application;
fig. 4 is a functional block diagram of a smart tv development performance testing apparatus according to an embodiment of the present application;
the achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The embodiment of the application provides a method for testing development performance of an intelligent television. The execution subject of the intelligent television development performance test method comprises, but is not limited to, at least one of a server, a terminal and the like which can be configured to execute the electronic equipment of the method provided by the embodiment of the application. In other words, the smart tv development performance test method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a method for testing development performance of an intelligent television according to an embodiment of the invention is shown. In this embodiment, the method for testing development performance of the smart television includes:
s1, shooting a test signal of an intelligent television to be detected by using a pre-corrected camera to obtain a primary test chart set, performing color detection on the intelligent television by using a preset color analyzer to obtain a color parameter set, and performing test signal recording on the intelligent television by using a pre-corrected recorder to obtain primary test audio.
In the embodiment of the invention, the pre-corrected camera refers to a camera which performs focal power adjustment, distortion correction and noise correction in advance, wherein the focal power adjustment refers to adjusting the focal point of the camera to be in the same plane with the intelligent television, the distortion correction refers to shooting a calibration plate by using the camera with the focal power adjusted to obtain a calibration plate picture, constructing a distortion formula according to the relation between each pixel on the calibration plate picture and the offset of the calibration plate, and performing distortion correction on the camera by using the distortion formula; the noise correction is to shoot a solid-color plate by using a camera after distortion correction to obtain a solid-color plate photo, determine the lens noise of the camera according to the solid-color plate photo, and correct the noise of the camera by using the lens noise.
In the embodiment of the invention, the step of shooting test signals of the smart television to be detected by using the pre-corrected camera to obtain a primary test atlas refers to the steps of enabling the smart television to be detected to play a preset test video, shooting the smart television playing the test video at intervals by using the pre-corrected camera to obtain primary test pictures in different time periods, and integrating all the primary test pictures into the primary test atlas.
In detail, the color detection of the smart tv by using a preset color analyzer, to obtain a color parameter set, includes:
performing global brightness detection on the intelligent television by using a preset color analyzer to obtain a regional brightness set;
extracting peak brightness and average brightness from the regional brightness set, calculating a brightness variance corresponding to the regional brightness set according to the average brightness, and taking the brightness variance as brightness uniformity;
performing global contrast detection on the intelligent television by using the color analyzer to obtain a region contrast set, and taking the average value of the region contrast set as a standard contrast;
performing multi-angle chromaticity detection on the intelligent television by using the color analyzer to obtain a chromaticity parameter set;
And collecting the peak brightness, the average brightness, the brightness uniformity, the standard contrast and the chromaticity parameter set into a chromaticity parameter set.
In detail, the step of performing global brightness detection on the smart tv by using a preset color analyzer, and obtaining a regional brightness set refers to performing brightness measurement on each region of the smart tv by using the color analyzer, and collecting brightness values of each region into a regional brightness set.
Specifically, the peak luminance refers to a region luminance having the largest luminance value in the region luminance set, and the average luminance refers to an average value of the luminance values in the region luminance set.
Specifically, the step of performing global contrast detection on the smart tv by using the color analyzer, and the step of obtaining a region contrast set refers to performing contrast test on each region of the smart tv, so as to integrate the contrast of each region into a region contrast set.
In detail, the step of performing multi-angle chromaticity detection on the smart tv by using the color analyzer, and the step of obtaining the chromaticity parameter set refers to performing chromaticity detection on the smart tv displaying various colors at various angles, so as to obtain the chromaticity parameter set.
In detail, the recording the test signal of the smart television by using the pre-corrected recorder to obtain the primary test audio means recording a mute environment by using the recorder to obtain a reference test audio, recording the smart television playing the test audio by using the recorder to obtain a default test audio, and denoising the default test audio by using the reference test audio to obtain the primary test audio.
In the embodiment of the invention, the primary test chart set is obtained by shooting the test signal of the intelligent television to be detected by utilizing the pre-corrected camera, and the picture quality of the primary test chart set can be ensured, so that the accuracy of identifying the subsequent screen defects is improved, the primary color display performance of the intelligent television can be measured by utilizing the preset color analyzer to carry out color detection on the intelligent television to obtain the color parameter set, the primary test audio is obtained by utilizing the pre-corrected recorder to carry out test signal recording on the intelligent television, and the background noise in the audio of the intelligent television can be removed, so that the identification accuracy of the subsequent distorted sounding defects is improved.
S2, splitting the primary test chart set into a screen-extinguishing test chart set and a bright screen test chart set, and sequentially carrying out image denoising, edge detection, image segmentation and image enhancement on each screen-extinguishing test image in the screen-extinguishing test chart set to obtain a secondary screen-extinguishing chart set.
In the embodiment of the invention, the screen-extinguishing test atlas refers to an atlas composed of pictures of the intelligent television in a plurality of screen-extinguishing states, and the bright screen test atlas refers to an atlas composed of pictures of the intelligent television in a plurality of Zhang Liangbing states.
In the embodiment of the present invention, splitting the primary test atlas into a screen-extinguishing test atlas and a screen-brightening test atlas includes:
selecting primary test pictures in the primary test picture set one by one as target primary pictures, and carrying out graying operation on the target primary pictures to obtain target gray-scale pictures;
calculating an average value of gray values of all pixels in the target gray picture, and taking the average value as a gray average value of the target gray picture;
judging whether the gray average value is larger than a preset gray threshold value or not;
if yes, adding the target primary picture corresponding to the gray average value into a preset bright screen test chart set;
If not, adding the target primary picture corresponding to the gray average value into a preset screen-extinguishing test chart set.
In detail, the target primary picture can be subjected to graying operation by using a cv2.cvtClolor function of OpenCV to obtain a target gray picture, and the gray threshold is obtained by experimental calculation in advance.
In the embodiment of the present invention, referring to fig. 2, the performing operations of picture denoising, edge detection, picture segmentation and picture enhancement on each screen-extinguishing test picture in the screen-extinguishing test picture set sequentially to obtain a secondary screen-extinguishing picture set includes:
s21, selecting screen-extinguishing test pictures in the screen-extinguishing test picture set one by one as target screen-extinguishing pictures, and performing median filtering operation on the target screen-extinguishing pictures to obtain denoising screen-extinguishing pictures;
s22, performing edge detection on the denoising screen picture by using a preset weight-cutting fitting algorithm to obtain a screen fitting edge;
s23, carrying out picture segmentation on the denoising screen-extinguishing picture according to the screen fitting edge to obtain a segmented screen-extinguishing picture;
s24, calculating a screen-extinguishing gray level histogram of the segmented screen-extinguishing picture, carrying out picture enhancement on the segmented screen-extinguishing picture by using the screen-extinguishing gray level histogram to obtain a secondary screen-extinguishing picture, and collecting all the secondary screen-extinguishing pictures into a secondary screen-extinguishing picture set.
In detail, the median filtering operation can be performed on the target screen-extinguishing picture by using the cv2.medianBlur function of the openCV, so as to obtain the denoising screen-extinguishing picture.
In detail, the performing edge detection on the denoising and screen-extinguishing picture by using a preset weight-cutting fitting algorithm to obtain a screen fitting edge includes:
extracting a primary screen edge set from the denoising screen picture by using a preset edge erosion algorithm;
fitting each primary screen edge in the primary screen edge set by using the following weight-loss fitting algorithm to obtain a standard fitting edge set:
wherein epsilon refers to a fitting error, R refers to the R-th pixel point corresponding to the primary screen edge, R refers to the total number of the pixel points corresponding to the primary screen edge, and w r Refer to the clipping weight of the r pixel point, ρ, σ is the slope coefficient of the fitted primary screen edge, γ is the displacement coefficient of the fitted primary screen edge, a r Refers to the abscissa, b, of the (r) th pixel point corresponding to the edge of the primary screen r The ordinate of the r pixel point corresponding to the primary screen edge is referred, and lambda refers to the fitting coefficient of the weight-cut fitting algorithm;
And splicing all standard fitting edges in the standard fitting edge set into a screen fitting edge.
In detail, the edge erosion algorithm may be a sobel operator or a canny operator, and the standard fitting edge set is obtained by fitting each primary screen edge in the primary screen edge set by using the weight-cut fitting algorithm, so that the situation that the fitting error is increased due to the short distance during fitting can be reduced, and the accuracy of edge fitting is improved.
In detail, the performing image segmentation on the denoising screen-extinguishing picture according to the screen fitting edge, and obtaining the segmentation screen-extinguishing picture refers to taking a part of the denoising screen-extinguishing picture, which is positioned inside the screen fitting edge, as the segmentation screen-extinguishing picture.
In the embodiment of the invention, the calculating of the screen-extinguishing gray level histogram of the divided screen-extinguishing picture refers to calculating a probability density function of various gray levels in the divided screen-extinguishing picture, generating the screen-extinguishing gray level histogram of the divided screen-extinguishing picture according to the probability density function, and performing gray level equalization operation on the divided screen-extinguishing picture by using the screen-extinguishing gray level histogram to obtain a secondary screen-extinguishing picture.
In the embodiment of the invention, the primary test chart set is split into the screen-extinguishing test chart set and the screen-brightening test chart set, so that the defect problem of the intelligent television during screen-extinguishing and screen-brightening can be detected respectively, and the secondary screen-extinguishing chart set is obtained by sequentially carrying out the operations of picture denoising, edge detection, picture segmentation and picture enhancement on each screen-extinguishing test picture in the screen-extinguishing test chart set, so that the defect on the intelligent television is more obvious, the subsequent feature extraction and feature identification are facilitated, and the test precision is improved.
S3, performing picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set, extracting features of secondary screen-extinguishing graph block sets in the secondary screen-extinguishing graph block set one by one to obtain screen-extinguishing feature sets, performing picture feature recognition on the screen-extinguishing feature sets to obtain a screen-extinguishing defect set, performing region feature recognition on the bright screen test graph set to obtain a bright screen defect set, and updating the bright screen defect set according to the screen-extinguishing defect set to obtain a standard screen defect set.
In an embodiment of the present invention, the secondary screen-extinguishing tile group set is a set composed of a plurality of secondary screen-extinguishing tile groups, each of the secondary screen-extinguishing tile groups is a combination composed of a plurality of secondary screen-extinguishing tiles, and each of the secondary screen-extinguishing tiles is a part of tiles in the secondary screen-extinguishing picture.
In the embodiment of the present invention, referring to fig. 3, the performing a picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set includes:
s31, carrying out inclination correction and picture lifting operation on each secondary screen-extinguishing picture in the secondary screen-extinguishing picture set in sequence to obtain a standard screen-extinguishing picture set;
s32, selecting standard screen-extinguishing pictures in the standard screen-extinguishing picture set one by one as target standard screen-extinguishing pictures, sequentially performing quadtree block, trigeminal tree block and binary tree block operation on the target standard screen-extinguishing pictures to obtain standard screen-extinguishing picture block groups, and collecting all the standard screen-extinguishing picture block groups into a standard screen-extinguishing picture block group set;
s33, selecting standard screen-extinguishing pattern blocks in the standard screen-extinguishing pattern block group one by one as a target standard screen-extinguishing pattern block group, and selecting standard screen-extinguishing pattern blocks in the target standard screen-extinguishing pattern block group one by one as target standard screen-extinguishing pattern blocks;
s34, adding the target standard screen-extinguishing block into a preset primary screen-extinguishing block group, adding standard screen-extinguishing blocks corresponding to the target standard screen-extinguishing block position in each standard screen-extinguishing block group except the target standard screen-extinguishing block group into the primary screen-extinguishing block group, obtaining a secondary screen-extinguishing block group, and collecting all secondary screen-extinguishing block groups into a secondary screen-extinguishing block group set.
In detail, the quadtree partitioning refers to dividing the target standard screen-extinguishing picture into four tiles with the same size, the trigeminal tree partitioning refers to dividing the tiles obtained by the quadtree partitioning into three tiles with the same size, and the binary tree partitioning refers to dividing the tiles obtained by the trigeminal tree partitioning into two tiles with the same size.
In the embodiment of the invention, the step of extracting the features of each secondary screen-extinguishing tile group in the secondary screen-extinguishing tile group one by one to obtain a screen-extinguishing feature group refers to the step of selecting each secondary screen-extinguishing tile group as a target secondary screen-extinguishing tile group one by one, and extracting the features of each secondary screen-extinguishing tile in the target secondary screen-extinguishing tile group by using a preset multi-stage convolution layer to obtain a target screen-extinguishing feature set, merging each target screen-extinguishing feature in the target screen-extinguishing feature set into a standard screen-extinguishing feature, and collecting all the standard screen-extinguishing features into the standard screen-extinguishing feature group.
In detail, the step of identifying the picture features of the screen-extinguishing feature group to obtain a screen-extinguishing defect set includes:
selecting standard screen-extinguishing characteristics in the screen-extinguishing characteristic group one by one as target standard screen-extinguishing characteristics, and calculating primary screen-extinguishing vectors corresponding to the target standard screen-extinguishing characteristics by using the following screen-extinguishing vector algorithm:
Wherein Z is the primary screen-extinguishing vector, K is the number of feature vectors of the target standard screen-extinguishing feature, K is the kth, x k Feature vectors representing the kth dimension in the target standard screen feature, alpha, beta and gamma are preset weight matrixes of the target standard screen feature,refers to x k Dimension of beta amount->Is the normalized coefficient of the screen-off vector algorithm, T is the transposed function;
and carrying out vector transcoding on the primary screen-off vectors to obtain primary screen-off semantics, converging all primary screen-off semantics into a screen-off semantic set, and screening out defect semantics from the screen-off semantic set to be converged into a screen-off defect set.
In the embodiment of the invention, the primary screen-off vector corresponding to the target standard screen-off feature is calculated by utilizing the screen-off vector algorithm, so that the target standard screen-off feature can be subjected to self-attention feature transcoding, and the semantic recognition efficiency is improved.
In detail, the screen-off defect set includes screen defects in various screen-off conditions and corresponding positions, such as screen scratches, screen breaks, and the screen-on defect set includes screen defects in various screen-on conditions and corresponding positions, such as screen light leakage, liquid crystal dead spots, and the like.
Specifically, the performing area feature recognition on the bright screen test atlas to obtain a bright screen defect set includes:
sequentially carrying out image denoising, edge detection, image segmentation and image enhancement on each bright screen test image in the bright screen test image set to obtain a secondary bright screen image set;
performing picture blocking operation on the secondary bright screen image set to obtain a secondary bright screen image block set;
and extracting the characteristics of each secondary bright screen block group in the secondary bright screen block group one by one to obtain a bright screen characteristic group, and identifying the picture characteristics of the bright screen characteristic group to obtain a bright screen defect set.
Specifically, the method for sequentially performing the operations of denoising, edge detection, image segmentation and image enhancement on each bright screen test image in the bright screen test image set to obtain a secondary bright screen image set is consistent with the method for sequentially performing the operations of denoising, edge detection, image segmentation and image enhancement on each bright screen test image in the bright screen test image set to obtain a secondary bright screen image set in the step S2, performing the image blocking operation on the secondary bright screen image set to obtain a secondary bright screen image block set, performing the feature extraction on each secondary bright screen image block set in the secondary bright screen image block set one by one, performing the image blocking operation on the secondary bright screen image set in the step S3 to obtain a secondary bright screen image block set, performing the feature extraction on each secondary bright screen image block set in the secondary bright screen image block set one by one to obtain a bright screen image feature set, and performing the image feature recognition on the bright screen feature set to obtain a bright screen defect set, where the method is not repeated.
In detail, updating the bright screen defect set according to the screen-off defect set to obtain a standard screen defect set, namely selecting screen-off defects in the screen-off defect set one by one as target screen-off defects, judging whether the defect positions of the target screen-off defects are the same as the positions of the bright screen defects in the bright screen defect set, if so, updating the defect types of the bright screen defects in the bright screen defect set at the same defect positions by using the defect types marked in the screen-off defect set; if not, adding the target screen-extinguishing defect into the bright screen defect set; and taking the updated bright screen defect set as a standard screen defect set until the target screen defect is the last screen defect in the screen defect set.
In the embodiment of the invention, the secondary screen-extinguishing picture block set is obtained by carrying out picture blocking operation on the secondary screen-extinguishing picture set, and the secondary screen-extinguishing picture set can be thinned, so that the interest area is more conveniently determined, the accuracy of defect identification is improved, the bright screen defect set is updated according to the screen-extinguishing defect set to obtain the standard screen defect set, and the screen defects in the bright screen state and the screen-extinguishing state can be respectively identified, thereby improving the accuracy of screen defect identification.
S4, carrying out band-pass filtering, amplitude adjustment and power amplification on the primary test audio in sequence to obtain secondary test audio, splitting the secondary test audio into a secondary audio sequence by utilizing a preset audio window, carrying out sound wave reconstruction on the secondary audio sequence to obtain standard audio characteristics, and carrying out audio characteristic identification on the standard audio characteristics to obtain a distortion pronunciation defect set.
In the embodiment of the invention, the secondary test audio is obtained by sequentially carrying out band-pass filtering, amplitude adjustment and power amplification on the primary test audio, so that the audio details can be further improved, and the accuracy of the subsequent audio feature recognition is improved.
In the embodiment of the present invention, the performing bandpass filtering, amplitude adjustment and power amplification on the primary test audio in sequence to obtain a secondary test audio includes:
carrying out band-pass filtering on the primary test audio by using a weighted recursive average filtering method to obtain primary filtered audio;
sequentially carrying out amplitude limiting filtering and jitter elimination filtering on the primary filtering audio to obtain secondary filtering audio;
and performing audio power amplification operation on the secondary filtering audio to obtain secondary test audio.
In detail, the audio window may be a hamming window, a raised cosine window, or a second order raised cosine window.
In the embodiment of the present invention, the performing acoustic wave reconstruction on the secondary audio sequence to obtain standard audio features includes:
selecting secondary audio in the secondary audio sequence one by one as target secondary audio, and performing audio decomposition on the target secondary audio to obtain a target audio decomposition tree;
and performing audio reconstruction on the target audio decomposition tree by using the following parity reconstruction formula to obtain a target audio reconstruction tree:
wherein (1)>Means +.f in the target audio reconstruction tree>The corresponding audio signal value when the abscissa of the mth tree node in the layer is i, e being the filter coefficient, ">Is a low-pass filter function, ">Is a high pass filter function, ">Refers to the objectCorresponding audio signal value when the abscissa of the 2m tree node in the j-th layer in the audio decomposition tree is i,/and>refers to the +.sup.th in the j-th layer in the target audio decomposition tree>Corresponding audio signal values when the abscissa of each tree node is i;
and extracting sample entropy features of the target audio reconstruction tree to obtain reconstructed audio features, and collecting all the reconstructed audio features into standard audio features.
In detail, the target secondary audio may be subjected to audio decomposition by using a wavelet packet analysis method or a mallat algorithm to obtain a target audio decomposition tree, and the sample entropy feature extraction is performed on the target audio reconstruction tree to obtain a reconstructed audio feature, which means that an audio sample entropy corresponding to the target audio reconstruction tree is calculated, and the audio sample entropy is used as the reconstructed audio feature.
In the embodiment of the invention, the target audio decomposition tree is obtained by carrying out audio reconstruction on the target audio decomposition tree by utilizing the parity reconstruction formula, and the audio denoising can be realized by low-pass filtering and high-pass filtering, so that more audio characteristic details are reserved.
In detail, the trained support vector machine model can be utilized to perform audio feature recognition on the standard audio features to obtain a distorted pronunciation defect set.
In the embodiment of the invention, the secondary test audio is obtained by sequentially carrying out band-pass filtering, amplitude adjustment and power amplification on the primary test audio, so that the audio detail can be further improved, the accuracy of the subsequent audio feature recognition is improved, the standard audio feature is obtained by carrying out sonic reconstruction on the secondary audio sequence, the audio feature recognition is carried out on the standard audio feature, the distortion pronunciation defect set is obtained, and the high-frequency detail feature and the low-frequency detail feature of the reserved audio can be improved, so that the recognition accuracy of true pronunciation defects is improved.
S5, performing color gamut labeling on the color parameter set to obtain a color detection result, performing gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result, and collecting the color detection result, the pronunciation detection result and the standard screen defect set into a performance test result.
In the implementation of the present invention, the performing color gamut labeling on the color parameter set to obtain a color detection result refers to generating a color gamut diagram of the smart television according to various color parameters in the color parameter set, and using the color gamut diagram, the color parameters and the color detection result as color detection results.
In detail, the performing a gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result refers to judging a distorted audio frequency band of the smart television according to distorted pronunciation defects in the distorted pronunciation defect set, so as to generate a pronunciation detection result.
In the embodiment of the invention, the performance test can be performed on the intelligent television on the three sides of screen color, screen defect and audio defect by collecting the color detection result, the pronunciation detection result and the standard screen defect set into the performance test result, so that the accuracy of the performance test is improved.
According to the embodiment of the invention, the primary test chart set is obtained by shooting test signals of the intelligent television to be detected by utilizing the pre-corrected camera, so that the picture quality of the primary test chart set can be ensured, the defect recognition precision of a subsequent screen can be improved, the primary color display performance of the intelligent television can be measured by utilizing the pre-corrected camera to obtain the color parameter set by carrying out color detection on the intelligent television, the primary test audio is obtained by utilizing the pre-corrected recorder to carry out test signal recording on the intelligent television, the background noise in the audio of the intelligent television can be removed, the recognition precision of the subsequent distortion pronunciation defects can be improved, the defect problems of the intelligent television in the process of the screen-extinguishing and the screen-lighting can be respectively detected by splitting the primary test chart set into the screen-extinguishing test chart set and the screen-lighting test chart set, the picture denoising, the edge detection, the picture segmentation and the picture enhancement operation can be carried out on each screen-extinguishing test chart in sequence, the defect on the intelligent television can be more obvious, the subsequent characteristic extraction and the characteristic extraction of the secondary picture can be carried out, the defect recognition precision can be further improved, the defect recognition precision can be carried out on the screen-extinguishing test chart set can be carried out, the screen-extinguishing defect can be further, the defect recognition precision can be further improved, and the defect can be respectively recognized by carrying out the screen-extinguishing test conditions can be respectively, and the defect detection can be more convenient, and the defect can be more convenient to be detected.
The secondary test audio is obtained by sequentially carrying out band-pass filtering, amplitude adjustment and power amplification operation on the primary test audio, so that the audio details can be further improved, the accuracy of the subsequent audio feature recognition is improved, the standard audio features are obtained by carrying out sonic reconstruction on the secondary audio sequence, the standard audio features are subjected to audio feature recognition to obtain a distorted pronunciation defect set, the high-frequency detail features and the low-frequency detail features of the reserved audio can be improved, the recognition accuracy of true pronunciation defects is improved, and the intelligent television can be subjected to performance test on three sides of screen colors, screen defects and audio defects by collecting the color detection results, the pronunciation detection results and the standard screen defect set into performance test results, so that the accuracy of performance test is improved. Therefore, the intelligent television development performance testing method provided by the invention can solve the problem of lower accuracy in performance testing of the intelligent television.
Fig. 4 is a functional block diagram of a smart tv development performance testing apparatus according to an embodiment of the present invention.
The intelligent television development performance testing device 100 can be installed in electronic equipment. According to the implemented functions, the smart tv development performance testing apparatus 100 may include a data acquisition module 101, a picture enhancement module 102, a screen test module 103, an audio test module 104, and a result generation module 105. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the data acquisition module 101 is configured to perform test signal shooting on an intelligent television to be detected by using a pre-corrected camera to obtain a primary test chart set, perform color detection on the intelligent television by using a preset color analyzer to obtain a color parameter set, and perform test signal recording on the intelligent television by using a pre-corrected recorder to obtain primary test audio;
the picture enhancement module 102 is configured to split the primary test atlas into a screen-extinguishing test atlas and a bright screen test atlas, and sequentially perform picture denoising, edge detection, picture segmentation and picture enhancement operations on each screen-extinguishing test picture in the screen-extinguishing test atlas to obtain a secondary screen-extinguishing atlas;
the screen test module 103 is configured to perform a picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set, perform feature extraction on each secondary screen-extinguishing graph block set in the secondary screen-extinguishing graph block set one by one to obtain a screen-extinguishing feature set, perform picture feature recognition on the screen-extinguishing feature set to obtain a screen-extinguishing defect set, perform region feature recognition on the bright screen test graph set to obtain a bright screen defect set, and update the bright screen defect set according to the screen-extinguishing defect set to obtain a standard screen defect set, where performing picture feature recognition on the screen-extinguishing feature set to obtain a screen-extinguishing defect set includes: selecting standard screen-extinguishing characteristics in the screen-extinguishing characteristic group one by one as target standard screen-extinguishing characteristics, and calculating primary screen-extinguishing vectors corresponding to the target standard screen-extinguishing characteristics by using the following screen-extinguishing vector algorithm:
Wherein Z is the primary radicalThe stage screen-extinguishing vector, K refers to the number of the characteristic vectors of the target standard screen-extinguishing characteristics, K refers to the kth, x k Feature vectors representing the kth dimension in the target standard screen feature, alpha, beta and gamma are preset weight matrixes of the target standard screen feature,refers to x k Dimension of beta vector,/->Is the normalized coefficient of the screen-off vector algorithm, T is the transposed function; vector transcoding is carried out on the primary screen-off vectors to obtain primary screen-off semantics, all primary screen-off semantics are collected into a screen-off semantic set, and defect semantics are screened out from the screen-off semantic set to be collected into a screen-off defect set;
the audio test module 104 is configured to sequentially perform bandpass filtering, amplitude adjustment, and power amplification on the primary test audio to obtain a secondary test audio, split the secondary test audio into a secondary audio sequence by using a preset audio window, perform sonic reconstruction on the secondary audio sequence to obtain a standard audio feature, and perform audio feature recognition on the standard audio feature to obtain a distorted pronunciation defect set;
the result generating module 105 is configured to perform color gamut labeling on the color parameter set to obtain a color detection result, perform gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result, and aggregate the color detection result, the pronunciation detection result, and the standard screen defect set into an integration performance test result.
In detail, each module in the intelligent television development performance testing apparatus 100 in the embodiment of the present invention adopts the same technical means as the intelligent television development performance testing method described in fig. 1 to 3, and can generate the same technical effects, which is not described herein.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. Multiple units or means as set forth in the system embodiments may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims (10)

1. The intelligent television development performance testing method is characterized by comprising the following steps of:
s1: shooting a test signal of the intelligent television to be detected by using a pre-corrected camera to obtain a primary test chart set, performing color detection on the intelligent television by using a preset color analyzer to obtain a color parameter set, and recording the test signal of the intelligent television by using a pre-corrected recorder to obtain primary test audio;
s2: splitting the primary test chart set into a screen-extinguishing test chart set and a bright screen test chart set, and sequentially carrying out image denoising, edge detection, image segmentation and image enhancement on each screen-extinguishing test image in the screen-extinguishing test chart set to obtain a secondary screen-extinguishing chart set;
s3: performing picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set, performing feature extraction on secondary screen-extinguishing graph block sets in the secondary screen-extinguishing graph block set one by one to obtain screen-extinguishing feature sets, performing picture feature recognition on the screen-extinguishing feature sets to obtain screen-extinguishing defect sets, performing region feature recognition on the bright screen test graph set to obtain bright screen defect sets, and updating the bright screen defect sets according to the screen-extinguishing defect sets to obtain standard screen defect sets, wherein the performing picture feature recognition on the screen-extinguishing feature sets to obtain the screen-extinguishing defect sets comprises:
S31: selecting standard screen-extinguishing characteristics in the screen-extinguishing characteristic group one by one as target standard screen-extinguishing characteristics, and calculating primary screen-extinguishing vectors corresponding to the target standard screen-extinguishing characteristics by using the following screen-extinguishing vector algorithm:
wherein Z is the primary screen-extinguishing vector, K is the number of feature vectors of the target standard screen-extinguishing feature, K is the kth, x k The feature vector representing the kth dimension in the target standard screen feature, alpha, beta and gamma are preset weight matrixes of the target standard screen feature, and the alpha, beta and gamma are ∈>Refers to x k Dimension of beta vector,/->Is the normalized coefficient of the screen-off vector algorithm, T is the transposed function;
s32: vector transcoding is carried out on the primary screen-off vectors to obtain primary screen-off semantics, all primary screen-off semantics are collected into a screen-off semantic set, and defect semantics are screened out from the screen-off semantic set to be collected into a screen-off defect set;
s4: sequentially performing bandpass filtering, amplitude adjustment and power amplification operation on the primary test audio to obtain secondary test audio, splitting the secondary test audio into a secondary audio sequence by utilizing a preset audio window, performing sonic reconstruction on the secondary audio sequence to obtain standard audio characteristics, and performing audio characteristic recognition on the standard audio characteristics to obtain a distortion pronunciation defect set;
S5: and performing color gamut labeling on the color parameter set to obtain a color detection result, performing gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result, and collecting the color detection result, the pronunciation detection result and the standard screen defect set into a performance test result.
2. The method for testing development performance of a smart tv according to claim 1, wherein the performing color detection on the smart tv using a preset color analyzer to obtain a color parameter set includes:
performing global brightness detection on the intelligent television by using a preset color analyzer to obtain a regional brightness set;
extracting peak brightness and average brightness from the regional brightness set, calculating a brightness variance corresponding to the regional brightness set according to the average brightness, and taking the brightness variance as brightness uniformity;
performing global contrast detection on the intelligent television by using the color analyzer to obtain a region contrast set, and taking the average value of the region contrast set as a standard contrast;
performing multi-angle chromaticity detection on the intelligent television by using the color analyzer to obtain a chromaticity parameter set;
And collecting the peak brightness, the average brightness, the brightness uniformity, the standard contrast and the chromaticity parameter set into a chromaticity parameter set.
3. The method for testing development performance of intelligent television according to claim 1, wherein the splitting the primary test atlas into a rest screen test atlas and a bright screen test atlas comprises:
selecting primary test pictures in the primary test picture set one by one as target primary pictures, and carrying out graying operation on the target primary pictures to obtain target gray-scale pictures;
calculating an average value of gray values of all pixels in the target gray picture, and taking the average value as a gray average value of the target gray picture;
judging whether the gray average value is larger than a preset gray threshold value or not;
if yes, adding the target primary picture corresponding to the gray average value into a preset bright screen test chart set;
if not, adding the target primary picture corresponding to the gray average value into a preset screen-extinguishing test chart set.
4. The method for testing development performance of intelligent television according to claim 1, wherein sequentially performing operations of picture denoising, edge detection, picture segmentation and picture enhancement on each screen-extinguishing test picture in the screen-extinguishing test picture set to obtain a secondary screen-extinguishing picture set comprises:
Selecting the screen-extinguishing test pictures in the screen-extinguishing test picture set one by one as target screen-extinguishing pictures, and performing median filtering operation on the target screen-extinguishing pictures to obtain denoising screen-extinguishing pictures;
performing edge detection on the denoising and screen-extinguishing picture by using a preset weight-cutting fitting algorithm to obtain a screen fitting edge;
carrying out picture segmentation on the denoising screen-extinguishing picture according to the screen fitting edge to obtain a segmented screen-extinguishing picture;
and calculating a screen-extinguishing gray level histogram of the segmented screen-extinguishing picture, carrying out picture enhancement on the segmented screen-extinguishing picture by using the screen-extinguishing gray level histogram to obtain a secondary screen-extinguishing picture, and collecting all the secondary screen-extinguishing pictures into a secondary screen-extinguishing picture set.
5. The method for testing development performance of intelligent television according to claim 4, wherein the performing edge detection on the de-noised screen picture by using a preset weight-cut fitting algorithm to obtain a screen fitting edge comprises:
extracting a primary screen edge set from the denoising screen picture by using a preset edge erosion algorithm;
fitting each primary screen edge in the primary screen edge set by using the following weight-loss fitting algorithm to obtain a standard fitting edge set:
Wherein epsilon refers to fitting error, R refers to the R pixel point corresponding to the edge of the primary screen, and R refers to the primary screenTotal number of pixel points corresponding to edge, w r Refer to the clipping weight of the r pixel point, ρ, σ is the slope coefficient of the fitted primary screen edge, γ is the displacement coefficient of the fitted primary screen edge, a r Refers to the abscissa, b, of the (r) th pixel point corresponding to the edge of the primary screen r The ordinate of the r pixel point corresponding to the primary screen edge is referred, and lambda refers to the fitting coefficient of the weight-cut fitting algorithm;
and splicing all standard fitting edges in the standard fitting edge set into a screen fitting edge.
6. The method for testing development performance of intelligent television according to claim 1, wherein the performing a picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set comprises:
sequentially performing inclination correction and picture lifting operation on each secondary screen-extinguishing picture in the secondary screen-extinguishing picture set to obtain a standard screen-extinguishing picture set;
selecting standard screen-extinguishing pictures in the standard screen-extinguishing picture set one by one as target standard screen-extinguishing pictures, sequentially performing quadtree block, trigeminal tree block and binary tree block operation on the target standard screen-extinguishing pictures to obtain standard screen-extinguishing picture block groups, and collecting all the standard screen-extinguishing picture block groups into a standard screen-extinguishing picture block group set;
Selecting standard screen-extinguishing pattern blocks in the standard screen-extinguishing pattern block group one by one as a target standard screen-extinguishing pattern block group, and selecting standard screen-extinguishing pattern blocks in the target standard screen-extinguishing pattern block group one by one as target standard screen-extinguishing pattern blocks;
adding the target standard screen-extinguishing block diagram block into a preset primary screen-extinguishing block diagram block, adding standard screen-extinguishing block diagram blocks corresponding to the target standard screen-extinguishing block diagram block position in each standard screen-extinguishing block diagram block except the target standard screen-extinguishing block diagram block in the standard screen-extinguishing block diagram block set into the primary screen-extinguishing block diagram block to obtain a secondary screen-extinguishing block diagram block, and collecting all secondary screen-extinguishing block diagram blocks into a secondary screen-extinguishing block diagram block set.
7. The method for testing development performance of intelligent televisions according to claim 1, wherein said performing region feature recognition on said bright screen test atlas to obtain a bright screen defect set includes:
sequentially carrying out image denoising, edge detection, image segmentation and image enhancement on each bright screen test image in the bright screen test image set to obtain a secondary bright screen image set;
performing picture blocking operation on the secondary bright screen image set to obtain a secondary bright screen image block set;
And extracting the characteristics of each secondary bright screen block group in the secondary bright screen block group one by one to obtain a bright screen characteristic group, and identifying the picture characteristics of the bright screen characteristic group to obtain a bright screen defect set.
8. The method for testing development performance of intelligent television according to claim 1, wherein the sequentially performing bandpass filtering, amplitude adjustment and power amplification on the primary test audio to obtain the secondary test audio comprises:
carrying out band-pass filtering on the primary test audio by using a weighted recursive average filtering method to obtain primary filtered audio;
sequentially carrying out amplitude limiting filtering and jitter elimination filtering on the primary filtering audio to obtain secondary filtering audio;
and performing audio power amplification operation on the secondary filtering audio to obtain secondary test audio.
9. The method for testing development performance of a smart television according to claim 1, wherein performing sonic reconstruction on the secondary audio sequence to obtain standard audio features comprises:
selecting secondary audio in the secondary audio sequence one by one as target secondary audio, and performing audio decomposition on the target secondary audio to obtain a target audio decomposition tree;
And performing audio reconstruction on the target audio decomposition tree by using the following parity reconstruction formula to obtain a target audio reconstruction tree:
wherein (1)>Means +.f in the target audio reconstruction tree>The corresponding audio signal value, e referring to the filter coefficient,is a low-pass filter function, ">Is a high pass filter function, ">Means the corresponding audio signal value when the abscissa of the 2m tree node in the j-th layer in the target audio decomposition tree is i, < ->Refers to the +.sup.th in the j-th layer in the target audio decomposition tree>Corresponding audio signal values when the abscissa of each tree node is i;
and extracting sample entropy features of the target audio reconstruction tree to obtain reconstructed audio features, and collecting all the reconstructed audio features into standard audio features.
10. An intelligent television development performance testing device, which is characterized by comprising:
the data acquisition module is used for shooting test signals of the intelligent television to be detected by using a pre-corrected camera to obtain a primary test chart set, performing color detection on the intelligent television by using a preset color analyzer to obtain a color parameter set, and performing test signal recording on the intelligent television by using a pre-corrected recorder to obtain primary test audio;
The picture enhancement module is used for splitting the primary test image set into a screen-extinguishing test image set and a bright screen test image set, and sequentially carrying out picture denoising, edge detection, picture segmentation and picture enhancement on each screen-extinguishing test picture in the screen-extinguishing test image set to obtain a secondary screen-extinguishing image set;
the screen test module is used for carrying out picture blocking operation on the secondary screen-extinguishing graph set to obtain a secondary screen-extinguishing graph block set, carrying out feature extraction on secondary screen-extinguishing graph block sets in the secondary screen-extinguishing graph block set one by one to obtain screen-extinguishing feature sets, carrying out picture feature recognition on the screen-extinguishing feature sets to obtain screen-extinguishing defect sets, carrying out region feature recognition on the bright screen test graph set to obtain bright screen defect sets, and updating the bright screen defect sets according to the screen-extinguishing defect sets to obtain standard screen defect sets, wherein the step of carrying out picture feature recognition on the screen-extinguishing feature sets to obtain the screen-extinguishing defect sets comprises the following steps: selecting standard screen-extinguishing characteristics in the screen-extinguishing characteristic group one by one as target standard screen-extinguishing characteristics, and calculating primary screen-extinguishing vectors corresponding to the target standard screen-extinguishing characteristics by using the following screen-extinguishing vector algorithm: Wherein Z is the primary screen-extinguishing vector, K is the number of feature vectors of the target standard screen-extinguishing feature, K is the kth, x k The feature vector representing the kth dimension in the target standard screen feature, alpha, beta and gamma are preset weight matrixes of the target standard screen feature, and the alpha, beta and gamma are ∈>Refers to x k Dimension of beta vector,/->Is the normalized coefficient of the screen-off vector algorithm, T is the transposed function; vector transcoding is carried out on the primary screen-off vectors to obtain primary screen-off semantics, all primary screen-off semantics are collected into a screen-off semantic set, and defect semantics are screened out from the screen-off semantic set to be collected into a screen-off defect set;
the audio test module is used for sequentially carrying out band-pass filtering, amplitude adjustment and power amplification operation on the primary test audio to obtain secondary test audio, splitting the secondary test audio into secondary audio sequences by utilizing a preset audio window, carrying out sonic reconstruction on the secondary audio sequences to obtain standard audio characteristics, and carrying out audio characteristic recognition on the standard audio characteristics to obtain a distortion pronunciation defect set;
the result generation module is used for carrying out color gamut labeling on the color parameter set to obtain a color detection result, carrying out gamut labeling on the distorted pronunciation defect set to obtain a pronunciation detection result, and collecting the color detection result, the pronunciation detection result and the standard screen defect set into an integration performance test result.
CN202310474021.6A 2023-04-28 2023-04-28 Smart television development performance testing method and device Active CN116405661B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310474021.6A CN116405661B (en) 2023-04-28 2023-04-28 Smart television development performance testing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310474021.6A CN116405661B (en) 2023-04-28 2023-04-28 Smart television development performance testing method and device

Publications (2)

Publication Number Publication Date
CN116405661A CN116405661A (en) 2023-07-07
CN116405661B true CN116405661B (en) 2023-09-29

Family

ID=87016117

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310474021.6A Active CN116405661B (en) 2023-04-28 2023-04-28 Smart television development performance testing method and device

Country Status (1)

Country Link
CN (1) CN116405661B (en)

Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080026995A (en) * 2006-09-22 2008-03-26 조영환 Method for detecting dot defects in flat display panel
CN104200478A (en) * 2014-09-12 2014-12-10 广东财经大学 Low-resolution touch screen image defect detection method based on sparse representation
CN106157286A (en) * 2015-04-03 2016-11-23 研祥智能科技股份有限公司 Image processing method and screen light leak test method
CN106650770A (en) * 2016-09-29 2017-05-10 南京大学 Mura defect detection method based on sample learning and human visual characteristics
CN110736751A (en) * 2019-11-06 2020-01-31 上海御微半导体技术有限公司 surface defect detection method and device
CN110784370A (en) * 2019-10-08 2020-02-11 Oppo（重庆）智能科技有限公司 Method and device for testing equipment, electronic equipment and medium
CN111077166A (en) * 2018-10-19 2020-04-28 北京金山云网络技术有限公司 Defect detection method and device of liquid crystal display and terminal equipment
WO2021012735A1 (en) * 2019-07-25 2021-01-28 研祥智能科技股份有限公司 Method and system for detecting screen display defects
CN113781396A (en) * 2021-08-11 2021-12-10 深圳市鑫信腾科技股份有限公司 Screen defect detection method, device, equipment and storage medium
CN114170184A (en) * 2021-12-08 2022-03-11 四川启睿克科技有限公司 Product image anomaly detection method and device based on embedded feature vector
CN114519714A (en) * 2022-04-20 2022-05-20 中导光电设备股份有限公司 Method and system for judging smudgy defect of display screen
CN115330789A (en) * 2022-10-13 2022-11-11 潍坊歌尔电子有限公司 Screen defect detection method, device, equipment and readable storage medium
CN115390291A (en) * 2022-08-24 2022-11-25 歌尔科技有限公司 Screen display state detection method, device and system and readable storage medium
CN115552628A (en) * 2021-04-29 2022-12-30 京东方科技集团股份有限公司 Display panel, manufacturing method thereof and display device
CN115908774A (en) * 2023-01-10 2023-04-04 佰聆数据股份有限公司 Quality detection method and device of deformed material based on machine vision

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020067184A1 (en) * 2000-10-27 2002-06-06 Three-Five Systems, Inc. Method and apparatus for testing color sequential, near-to-the-eye, and similar display devices
US6734898B2 (en) * 2001-04-17 2004-05-11 General Instrument Corporation Methods and apparatus for the measurement of video quality
US20050286753A1 (en) * 2004-06-25 2005-12-29 Triant Technologies Inc. Automated inspection systems and methods
CN104392432A (en) * 2014-11-03 2015-03-04 深圳市华星光电技术有限公司 Histogram of oriented gradient-based display panel defect detection method
US10681344B2 (en) * 2017-12-15 2020-06-09 Samsung Display Co., Ltd. System and method for mura detection on a display
CN110414538B (en) * 2019-07-24 2022-05-27 京东方科技集团股份有限公司 Defect classification method, defect classification training method and device thereof

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080026995A (en) * 2006-09-22 2008-03-26 조영환 Method for detecting dot defects in flat display panel
CN104200478A (en) * 2014-09-12 2014-12-10 广东财经大学 Low-resolution touch screen image defect detection method based on sparse representation
CN106157286A (en) * 2015-04-03 2016-11-23 研祥智能科技股份有限公司 Image processing method and screen light leak test method
CN106650770A (en) * 2016-09-29 2017-05-10 南京大学 Mura defect detection method based on sample learning and human visual characteristics
CN111077166A (en) * 2018-10-19 2020-04-28 北京金山云网络技术有限公司 Defect detection method and device of liquid crystal display and terminal equipment
WO2021012735A1 (en) * 2019-07-25 2021-01-28 研祥智能科技股份有限公司 Method and system for detecting screen display defects
CN110784370A (en) * 2019-10-08 2020-02-11 Oppo（重庆）智能科技有限公司 Method and device for testing equipment, electronic equipment and medium
CN110736751A (en) * 2019-11-06 2020-01-31 上海御微半导体技术有限公司 surface defect detection method and device
CN115552628A (en) * 2021-04-29 2022-12-30 京东方科技集团股份有限公司 Display panel, manufacturing method thereof and display device
CN113781396A (en) * 2021-08-11 2021-12-10 深圳市鑫信腾科技股份有限公司 Screen defect detection method, device, equipment and storage medium
CN114170184A (en) * 2021-12-08 2022-03-11 四川启睿克科技有限公司 Product image anomaly detection method and device based on embedded feature vector
CN114519714A (en) * 2022-04-20 2022-05-20 中导光电设备股份有限公司 Method and system for judging smudgy defect of display screen
CN115390291A (en) * 2022-08-24 2022-11-25 歌尔科技有限公司 Screen display state detection method, device and system and readable storage medium
CN115330789A (en) * 2022-10-13 2022-11-11 潍坊歌尔电子有限公司 Screen defect detection method, device, equipment and readable storage medium
CN115908774A (en) * 2023-01-10 2023-04-04 佰聆数据股份有限公司 Quality detection method and device of deformed material based on machine vision

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Lingfeng Kong等.Detection of Water-Stains Defects in TFT-LCD Based on Machine Vision.《2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)，Beijing, China，13-15 October 2018》.2018,全文. *
曾毅.基于机器视觉的液晶屏点灯缺陷检测系统关键技术研究.《中国优秀硕士学位论文全文数据库（电子期刊）信息科技辑》.2018,全文. *

Also Published As

Publication number Publication date
CN116405661A (en) 2023-07-07

Similar Documents

Publication Publication Date Title
CN114584849B (en) Video quality evaluation method, device, electronic equipment and computer storage medium
US20110019094A1 (en) System and method for random noise estimation in a sequence of images
CN111709265A (en) Camera monitoring state classification method based on attention mechanism residual error network
CN112149476B (en) Target detection method, device, equipment and storage medium
CN110163807B (en) Low-illumination image enhancement method based on expected bright channel
CN110111269A (en) Low-light-level imaging algorithm and device based on multiple dimensioned context converging network
CN111460964A (en) Moving target detection method under low-illumination condition of radio and television transmission machine room
CN114022732A (en) Extremely dark light object detection method based on RAW image
CN112070682A (en) Method and device for compensating image brightness
CN116433496A (en) Image denoising method, device and storage medium
US6792142B1 (en) Image sensing system with histogram modification
CN113132700B (en) Method and system for adjusting contrast of projector
CN116777811A (en) Real scene image quality evaluation method based on multi-scale space pyramid pooling
CN112712482B (en) Image defogging method based on linear learning model
CN116405661B (en) Smart television development performance testing method and device
US11983853B1 (en) Techniques for generating training data for machine learning enabled image enhancement
CN110378271B (en) Gait recognition equipment screening method based on quality dimension evaluation parameters
CN116095291B (en) Image preprocessing method for media stream image transmission
CN113129300A (en) Drainage pipeline defect detection method, device, equipment and medium for reducing false detection rate
CN110020999B (en) Uncooled infrared thermal image self-adaptive mapping method based on homomorphic filtering
CN115170420A (en) Image contrast processing method and system
CN116630365A (en) Parabolic detection method and device, electronic equipment and storage medium
CN112995666B (en) Video horizontal and vertical screen conversion method and device combined with scene switching detection
WO2021189460A1 (en) Image processing method and apparatus, and movable platform
KR20220057024A (en) Fine dust detecting solution and system by computing brightness residual based on AI

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