CN115035515A - Nang identification method - Google Patents

Nang identification method Download PDF

Info

Publication number
CN115035515A
CN115035515A CN202210672381.2A CN202210672381A CN115035515A CN 115035515 A CN115035515 A CN 115035515A CN 202210672381 A CN202210672381 A CN 202210672381A CN 115035515 A CN115035515 A CN 115035515A
Authority
CN
China
Prior art keywords
image
thermal infrared
nang
visible light
features
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210672381.2A
Other languages
Chinese (zh)
Inventor
李超
殷光强
常益凡
杨晓宇
刘学婷
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.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
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 University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202210672381.2A priority Critical patent/CN115035515A/en
Publication of CN115035515A publication Critical patent/CN115035515A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/68—Food, e.g. fruit or vegetables
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/30—Noise filtering
    • 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
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
    • G06V10/464—Salient features, e.g. scale invariant feature transforms [SIFT] using a plurality of salient features, e.g. bag-of-words [BoW] representations
    • 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
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • 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
    • 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
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a Nang identification method, which comprises the following steps: a: collecting a visible light image and a thermal infrared image of the Nang, carrying out normalization processing on the visible light image, carrying out completion processing and clear processing on the lost characteristic of the thermal infrared image due to shielding, carrying out enhancement processing on edge details and textures on the thermal infrared image, and obtaining an enhanced thermal infrared image after processing; b: registering the enhanced thermal infrared image and the normalized visible light image; c: and performing super-resolution enhancement on the registered image, extracting SIFT features, LBP features and HOG features in the image, and identifying the type of the Nang based on the SIFT features, the LBP features and the HOG features and counting correspondingly. The method combines the visible light technology and the thermal infrared technology, can realize the accurate detection and the high-efficiency recognition of the Nang, and improves the speed and the accuracy of the Nang recognition.

Description

Nang identification method
Technical Field
The invention belongs to the technical field of Nang identification, and particularly relates to a Nang identification method which is mainly used for identifying and counting the types of Nang.
Background
The crusty pancake is a round cake baked by wheat flour or corn flour. The crusty pancakes are of various varieties, namely fifty or more, common crusty pancakes, oil crusty pancakes, pit crusty pancakes, sesame crusty pancakes, sliced crusty pancakes, Heerman crusty pancakes and the like, each crusty pancakes has different patterns, each pattern has different names, the shapes of the patterns are not completely the same, but the patterns mainly take a round shape and are different in size and thickness.
The Nang is mainly baked in a Nang pit, the Nang pit has three different specifications of big, middle and small, the height of the Nang pit is about 0.8 meter, the upper opening is small and about 60 centimeters, and the lower part is large and about 1.2 meters. Because the crusty pancake pit is large, when the crusty pancake is baked, a plurality of different types of crusty pancake are usually placed in the same crusty pancake pit for baking. And because the requirement of the crusty pancakes is large, the crusty pancakes of a large number are required to be baked every day. Therefore, the types of the crusty pancakes need to be identified and counted so as to be convenient for accurately knowing the types and the number of the prepared crusty pancakes and provide data support for the subsequent preparation of the crusty pancakes of the corresponding types and the number.
At present, the types of crusty pancakes are generally identified and counted manually, but the technical problems of high manual strength and easy error caused by negligence exist. In addition, the prior art also has more image recognition technologies, but the Nang is usually in an environment with poor light because of the special requirement of baking the Nang by using the Nang pit, so the prior image recognition technology can not be effectively applied to the Nang recognition.
For this reason, it is necessary to combine a more scientific image recognition technique to solve the above technical problem.
Disclosure of Invention
The invention aims to overcome the technical problems in the prior art and provides a Nang identification method, which combines a visible light technology and a thermal infrared technology, can realize the accurate detection and the efficient identification of the Nang and improves the identification speed and the accuracy of the Nang.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a Nang identification method is characterized by comprising the following steps:
step A: collecting a visible light image and a thermal infrared image of the Nang, carrying out normalization processing on the visible light image, carrying out completion processing and clear processing on the lost characteristics of the thermal infrared image due to shielding, carrying out enhancement processing on edge details and textures on the thermal infrared image, and obtaining an enhanced thermal infrared image after processing;
and B: registering the enhanced thermal infrared image and the normalized visible light image;
and C: and performing super-resolution enhancement on the registered image, extracting SIFT features, LBP features and HOG features in the image, and identifying the type of the Nang based on the SIFT features, the LBP features and the HOG features and counting correspondingly.
In the step A, the thermal infrared camera and the visible light camera are respectively utilized to collect the images of the Nang.
The thermal infrared camera and the visible light camera are respectively arranged above the crusty pancake pit.
In the step A, edge details and textures of the supplemented thermal infrared image are enhanced by a method based on an unsharp mask.
In the step A, the completion processing refers to performing completion on the lost features caused by shielding by adopting a local binary pattern analysis method, and the clear processing refers to improving the definition of the completion features by adopting an optimization algorithm enhanced by local features.
In the step B, the specific method of registration is:
s1: unifying the resolution of the thermal infrared image and the visible light image;
s2: registering the thermal infrared image and the visible light image by using a least square solution;
s3: and removing noise in the registered image by adopting a low-pass filter to obtain a standard registered image.
In step S1, the resolutions of the thermal infrared image and the visible light image are unified by using a bilinear interpolation method, and the resolution of the thermal infrared image and the resolution of the visible light image are 32 × 32 and 320 × 320 after the thermal infrared image and the visible light image are unified.
In the step C, performing super-resolution enhancement on the registered image means: and enhancing the super-resolution of the image by four times by adopting a Retinex-CNN algorithm.
By adopting the technical scheme, the invention has the beneficial technical effects that:
1. according to the invention, the enhancing technology of the thermal infrared image and the visible light image and the target recognition technology are adopted, and based on the combination of the two technologies, when the Nang is taken out from the Nang pit, the accurate detection of the Nang is realized through the enhancing technology of the thermal infrared image and the visible light image, then the efficient recognition of the Nang is realized through the target recognition technology, and the speed and the accuracy of the Nang recognition are effectively improved. Further, the specific advantages of each step are as follows:
in the step A, the influence of light ray change on the visible light image can be reduced through normalization processing, and the identification effect can be improved. The characteristic area in the thermal infrared image can be increased through the completion processing, the characteristics in the thermal infrared image can be clearer and more obvious through the clear processing and the enhancement processing, and the improvement of the identification speed, the identification effect and the identification precision is facilitated.
In the step B, the enhanced thermal infrared image and the normalized visible light image are registered, so that the characteristics of the visible light image and the characteristics of the thermal infrared image can be superposed, and the success rate of the subsequent image identification can be improved.
In the step C, super-resolution enhancement is performed on the registered image, and then the type of the Nang is identified and counted correspondingly through the extracted SIFT feature, LBP feature and HOG feature, so that the features in the image are clearer, and the accurate identification and detection of the Nang are further improved.
2. The method based on the unsharp mask is adopted to carry out enhancement processing on the edge details and the textures of the supplemented thermal infrared image, which is beneficial to further enhancing the remarkable characteristics in the image and provides support for subsequent effective and accurate identification.
3. The invention adopts a local binary pattern analysis method to carry out completion processing, adopts an optimization algorithm of local feature enhancement to carry out clear processing, and is beneficial to clearly completing the shielded part in the image.
4. When the invention is aligned, two images with fixed resolution can be obtained by unifying the resolution of the thermal infrared image and the visible light image; registering the thermal infrared image and the visible light image by using a least square solution, so that the characteristics of the visible light can be coincided with the thermal infrared characteristics; and a low-pass filter is adopted to remove noise in the registered image, so that noise interference is reduced. The combination of the three components is beneficial to the rapid processing of subsequent images and can increase the recognition success rate of the images.
5. Because the invention adopts the thermal infrared technology and the visible light technology, the types of the crusty pancakes can be effectively identified even in the environment with poor light, thereby being beneficial to saving electric energy.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
Example 1
The embodiment discloses a Nang identification method, which can realize accurate identification and counting of Nang by using an image identification technology when the Nang is prepared in large batch. As shown in fig. 1, it includes the following steps:
step A: because a single crusty pancake hole single can bake a plurality of different types of crusty pancake, and all need take out from the crusty pancake hole after baking, consequently, in order to facilitate image acquisition, this embodiment preferably sets up hot infrared camera and visible light camera respectively in the top of crusty pancake hole, when the crusty pancake is come out from the centre gripping in the crusty pancake hole, just can utilize hot infrared camera and visible light camera to gather the visible light image and the hot infrared image of crusty pancake respectively.
After collecting visible light image and hot infrared image, in order to reduce the influence of light change to the visible light image, and shelter from the influence that causes to hot infrared image when reducing the centre gripping naan, need carry out normalization processing to the visible light image, and carry out completion processing and clear processing to the characteristic that the hot infrared image loses because of sheltering from, afterwards, for the saliency of the characteristic in promoting hot infrared image, still need carry out the enhancement processing of edge detail and texture to hot infrared image again, obtain the hot infrared image of reinforcing after the processing. In order to improve the effect of completion processing, a local binary pattern analysis method is preferentially adopted to complete the lost features caused by shielding. In order to improve the effect of the definition processing, the definition of the completion feature is preferably improved by adopting an optimization algorithm of local feature enhancement. The local binary pattern analysis method and the local feature enhanced optimization algorithm are both conventional random image occlusion reconstruction algorithms, and therefore the specific processing process is not repeated. In order to improve the effect of the enhancement processing, it is preferable to perform edge detail and texture enhancement processing on the supplemented thermal infrared image by using a method based on an unsharp mask.
The crusty pancake is characterized in that when the crusty pancake is clamped out of the crusty pancake pit, the surface of the crusty pancake is partially shielded, the clamping position of the crusty pancake is different when the crusty pancake is taken out every time, and the thermal infrared image has larger change under different shielding conditions, so that in order to further improve the accuracy of crusty pancake identification, the shielded part of the crusty pancake in the thermal infrared image needs to be subjected to complete supplement processing and clear processing, so that the identification accuracy is improved.
And B: registering the enhanced thermal infrared image and the normalized visible light image, wherein the specific registering method comprises the following steps:
s1: and unifying the resolutions of the thermal infrared image and the visible light image by a bilinear interpolation method, wherein the resolution of the thermal infrared image is 32 x 32 after unification, and the resolution of the visible light image is 320 x 320.
S2: and registering the thermal infrared image and the visible light image by using a least square solution, wherein the registration mainly synthesizes the thermal infrared image and the visible light image into one image.
S3: and removing noise in the registered image by adopting a low-pass filter to obtain a standard registered image.
And C: performing super-resolution enhancement on the registered image by adopting a Retinex-CNN algorithm, and preferably enhancing the super-resolution of the registered image by four times; and extracting SIFT features, LBP features and HOG features in the super-resolution enhanced image, and then identifying the species of the naan based on the SIFT features, LBP features and HOG features and counting correspondingly.
Regarding the identification of the Nang, the improved thermal infrared image super-resolution enhancement algorithm Retinex-CNN algorithm is preferably used in the step, and the network of the Retinex-CNN algorithm is divided into two parts: the denoising enhancement network Enhance-Net of the reflection component is added to the reflection component decomposition network Decompose-Net of the low-illumination image reflection component and the illumination component constructed by the combined action of a plurality of convolution and activation layers in the 29 CNN network. And then, the prior knowledge of the near infrared image is utilized, and the identification performance is improved by extracting the features insensitive to the difference between the two modes according to an invariant feature extraction algorithm. And selecting three characteristics of SIFT, LBP and HOG to extract local characteristics of the registered images, and finally identifying the thermal infrared Nang images based on full-characteristic fusion to realize the rapid identification of the thermal infrared Nang images.
Example 2
This example verifies the method of example 1 as follows:
and (3) verifying conditions: in a factory, the illumination is 3lux, and 100 crusty pancakes with different types and specifications are tested by adopting crusty pancakes with the diameter of 1.5 m.
And (3) verification process: the workers clamp the baked crusty pancakes from the crusty pancakes pit, and the crusty pancakes are identified and counted by the invention through the visible light and the infrared camera.
And (4) verification result: compared with manual work, the counting accuracy reaches 100% and the recognition accuracy reaches 95% after the recognition is carried out by adopting the method.
In summary, the invention combines the visible light technology and the thermal infrared technology, realizes the accurate detection and the high-efficiency recognition of the Nang, and improves the recognition speed and the accuracy of the Nang.
While the invention has been described with reference to specific embodiments, any feature disclosed in this specification may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise; all of the disclosed features, or all of the method or process steps, may be combined in any combination, except mutually exclusive features and/or steps.

Claims (8)

1. A Nang identification method is characterized by comprising the following steps:
step A: collecting a visible light image and a thermal infrared image of the Nang, carrying out normalization processing on the visible light image, carrying out completion processing and clear processing on the lost characteristic of the thermal infrared image due to shielding, carrying out enhancement processing on edge details and textures on the thermal infrared image, and obtaining an enhanced thermal infrared image after processing;
and B: registering the enhanced thermal infrared image and the normalized visible light image;
and C: and performing super-resolution enhancement on the registered image, extracting SIFT features, LBP features and HOG features in the image, and identifying the type of the Nang based on the SIFT features, the LBP features and the HOG features and counting correspondingly.
2. The Nang identification method according to claim 1, characterized in that: in the step A, the images of the Nang are collected by a thermal infrared camera and a visible light camera respectively.
3. The Nang identification method according to claim 2, characterized in that: the thermal infrared camera and the visible light camera are respectively arranged above the crusty pancake pit.
4. The Nang identification method according to claim 1, characterized in that: in the step A, edge details and textures of the supplemented thermal infrared image are enhanced by a method based on an unsharp mask.
5. The Nang identification method according to claim 1, characterized in that: in the step A, the completion processing refers to performing completion on the lost features caused by shielding by adopting a local binary pattern analysis method, and the clear processing refers to improving the definition of the completion features by adopting an optimization algorithm enhanced by local features.
6. The method for identifying a crusty pancake according to any one of claims 1 to 5, wherein the method comprises the following steps: in step B, the specific method of registration is:
s1: unifying the resolution of the thermal infrared image and the visible light image;
s2: registering the thermal infrared image and the visible light image by using a least square solution;
s3: and removing noise in the registered image by adopting a low-pass filter to obtain a standard registered image.
7. The Nang identification method according to claim 6, characterized in that: in step S1, the thermal infrared image and the visible light image are unified by using a bilinear interpolation method, where the thermal infrared image and the visible light image have a resolution of 32 × 32 and the visible light image has a resolution of 320 × 320 after the thermal infrared image and the visible light image are unified.
8. The Nang identification method according to claim 1, characterized in that: in the step C, performing super-resolution enhancement on the registered image refers to: and enhancing the super-resolution of the image by four times by adopting a Retinex-CNN algorithm.
CN202210672381.2A 2022-06-15 2022-06-15 Nang identification method Pending CN115035515A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210672381.2A CN115035515A (en) 2022-06-15 2022-06-15 Nang identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210672381.2A CN115035515A (en) 2022-06-15 2022-06-15 Nang identification method

Publications (1)

Publication Number Publication Date
CN115035515A true CN115035515A (en) 2022-09-09

Family

ID=83125675

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210672381.2A Pending CN115035515A (en) 2022-06-15 2022-06-15 Nang identification method

Country Status (1)

Country Link
CN (1) CN115035515A (en)

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101701915A (en) * 2009-11-13 2010-05-05 江苏大学 Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision
CN106960202A (en) * 2017-04-11 2017-07-18 广西师范大学 A kind of smiling face's recognition methods merged based on visible ray with infrared image
CN108806375A (en) * 2018-05-09 2018-11-13 河南工学院 A kind of interactive teaching method and system based on image recognition
CN111898649A (en) * 2020-07-07 2020-11-06 西南林业大学 Wood tree species classification and identification method based on visible light or near infrared spectrum analysis
CN112733950A (en) * 2021-01-18 2021-04-30 湖北工业大学 Power equipment fault diagnosis method based on combination of image fusion and target detection
US20210272300A1 (en) * 2018-11-06 2021-09-02 Flir Commercial Systems, Inc. Automatic co-registration of thermal and visible image pairs
CN113643345A (en) * 2021-07-27 2021-11-12 数量级(上海)信息技术有限公司 Multi-view road intelligent identification method based on double-light fusion
US20220044374A1 (en) * 2019-12-17 2022-02-10 Dalian University Of Technology Infrared and visible light fusion method
CN114140366A (en) * 2021-11-01 2022-03-04 西安电子科技大学 Infrared image and visible light image fusion method, system, equipment and terminal
CN114331879A (en) * 2021-12-15 2022-04-12 中国船舶重工集团公司第七0九研究所 Visible light and infrared image registration method for equalized second-order gradient histogram descriptor
CN114511527A (en) * 2022-01-24 2022-05-17 安徽大学 Textile spinning cake forming defect detection method based on expanded local binary pattern

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101701915A (en) * 2009-11-13 2010-05-05 江苏大学 Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision
CN106960202A (en) * 2017-04-11 2017-07-18 广西师范大学 A kind of smiling face's recognition methods merged based on visible ray with infrared image
CN108806375A (en) * 2018-05-09 2018-11-13 河南工学院 A kind of interactive teaching method and system based on image recognition
US20210272300A1 (en) * 2018-11-06 2021-09-02 Flir Commercial Systems, Inc. Automatic co-registration of thermal and visible image pairs
US20220044374A1 (en) * 2019-12-17 2022-02-10 Dalian University Of Technology Infrared and visible light fusion method
CN111898649A (en) * 2020-07-07 2020-11-06 西南林业大学 Wood tree species classification and identification method based on visible light or near infrared spectrum analysis
CN112733950A (en) * 2021-01-18 2021-04-30 湖北工业大学 Power equipment fault diagnosis method based on combination of image fusion and target detection
CN113643345A (en) * 2021-07-27 2021-11-12 数量级(上海)信息技术有限公司 Multi-view road intelligent identification method based on double-light fusion
CN114140366A (en) * 2021-11-01 2022-03-04 西安电子科技大学 Infrared image and visible light image fusion method, system, equipment and terminal
CN114331879A (en) * 2021-12-15 2022-04-12 中国船舶重工集团公司第七0九研究所 Visible light and infrared image registration method for equalized second-order gradient histogram descriptor
CN114511527A (en) * 2022-01-24 2022-05-17 安徽大学 Textile spinning cake forming defect detection method based on expanded local binary pattern

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
S. DIVYA MEENA等: "Smart Animal Detection and Counting Framework for Monitoring Livestock in an Autonomous Unmanned Ground Vehicle Using Restricted Supervised Learning and Image Fusion", 《NEURAL PROCESSING LETTERS》, vol. 53, 13 February 2021 (2021-02-13), pages 1253, XP037423233, DOI: 10.1007/s11063-021-10439-4 *
宗泽等: "基于多尺度滤波和稀疏表示的图像融合方法研究", 《农业工程》, vol. 7, no. 6, 30 November 2017 (2017-11-30), pages 138 - 22 *
杨万利: "基于红外和可见光图像融合技术的苹果早期瘀伤检测研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 1, 15 December 2011 (2011-12-15), pages 138 - 1701 *
胡建文: "基于多尺度滤波和稀疏表示的图像融合方法研究", 《中国博士学位论文全文数据库 信息科技辑》, no. 9, 15 September 2014 (2014-09-15), pages 138 - 22 *

Similar Documents

Publication Publication Date Title
CN109447945B (en) Quick counting method for basic wheat seedlings based on machine vision and graphic processing
CN111126174A (en) Visual detection method for robot to grab parts
WO2020224458A1 (en) Method for detecting corona discharge employing image processing
CN112347887B (en) Object detection method, object detection device and electronic equipment
CN106067026A (en) A kind of Feature extraction and recognition search method of microimage of Chinese medical herb
CN105930854A (en) Manipulator visual system
CN110110131B (en) Airplane cable support identification and parameter acquisition method based on deep learning and binocular stereo vision
CN109241983A (en) A kind of cigarette image-recognizing method of image procossing in conjunction with neural network
CN111251336A (en) Double-arm cooperative intelligent assembly system based on visual positioning
CN110640741A (en) Grabbing industrial robot with regular-shaped workpiece matching function
CN108960100A (en) A kind of recognition methods of the sugarcane sugarcane section based on image procossing
WO2023231262A1 (en) Hoisting wire rope tension measurement method based on visual vibration frequency identification
CN111008642A (en) High-resolution remote sensing image classification method and system based on convolutional neural network
CN114332081B (en) Textile surface abnormity determination method based on image processing
CN116109637A (en) System and method for detecting appearance defects of turbocharger impeller based on vision
CN115108466A (en) Intelligent positioning method for container spreader
CN115035515A (en) Nang identification method
CN108269264B (en) Denoising and fractal method of bean kernel image
CN105787955A (en) Sparse segmentation method and device of strip steel defect
CN112991327A (en) Steel grid welding system and method based on machine vision and terminal equipment
CN111815542A (en) Tree ring image medullary center positioning and ring measuring method
CN113744263B (en) Method for rapidly detecting surface defects of small-size mosaic ceramic
CN114266975B (en) Litchi fruit detection and counting method for unmanned aerial vehicle remote sensing image
CN110189301B (en) Foreign matter detection method for generator stator core steel sheet stacking platform
CN106682617B (en) A kind of image clearly based on frequency spectrum segment information and fuzzy discrimination feature extracting method

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