CN103344647B - The defect inspection method of a kind of potato - Google Patents
The defect inspection method of a kind of potato Download PDFInfo
- Publication number
- CN103344647B CN103344647B CN201310280939.3A CN201310280939A CN103344647B CN 103344647 B CN103344647 B CN 103344647B CN 201310280939 A CN201310280939 A CN 201310280939A CN 103344647 B CN103344647 B CN 103344647B
- Authority
- CN
- China
- Prior art keywords
- potato
- pixel
- thermal imagery
- temperature value
- thermal
- 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.)
- Expired - Fee Related
Links
Landscapes
- Apparatuses For Bulk Treatment Of Fruits And Vegetables And Apparatuses For Preparing Feeds (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses the defect inspection method of a kind of potato, utilize thermal imaging to reach the peeled potatoes identifying fast and damage, rot and germinate; The thermal imagery processing procedure of meter place machine is: carry out medium filtering or mean filter denoising to thermal imagery, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, does is the absolute value of calculated difference greater than the area S of the corresponding material object of pixel of 1 DEG C, if S is greater than 25? mm
2, be then defective potato.The defect inspection method efficiency of potato of the present invention is high, identification accuracy is good, can be applied to the defects detection field of potato.
Description
Technical field
The present invention relates to a kind of method for quick identification of Potato Quality, particularly relate to a kind of method utilizing thermal imaging method to differentiate defect potato fast, belong to field of non destructive testing.
Background technology
Potato is one of four generalized grain crops of human society now, is only second to paddy rice, corn and wheat.Potato contains a large amount of starch, can provide the abundant nutrition energy for eater.Along with the fast development of food industry, the edible way of potato is more and more diversified, and the demand such as mashed potatoes, Hash Browns based on potato full-powder is increasing.
Potato full-powder have fresh potato cook after local flavor, nutritive loss is few, quality stability good, the easy to process and advantage such as easy that is shaped, so be widely used in compound French fries/sheet, facilitate mashed potato, fast food (or quick-frozen), expanded and baby food etc., be multiple casual health raw-food material popular in the world.At present, the production of potato full-powder all adopts mechanization, compared with artificial powder process, raises the efficiency greatly.
In potato full-powder preparation process, it is first decortication.Usually, the affinity between epidermis and meat is comparatively strong, and both fit tightly, and is difficult to be separated.Peeling method has three kinds, comprises chemical peeling, grinding-wheel grinder and steam peeling.
Chemical peeling method cost improves, and can pollute potato meat and environment, now no longer adopt.
Grinding-wheel grinder pico farad be application wheel to the direct grinding of potato, it is large to there is meat damage in this method, and the shortcomings such as manufactured goods yield is lower, cause serious waste.
Steam peeling steams skin to potato class as the fruits such as potato carry out high-temperature instantaneous, and epidermis is expanded, separation rapidly, to meat not damaged, and has high, the free of contamination advantage of decortication efficiency, has become the main stream approach of decortication, and provide basis for obtaining high yield potato powder.
Potato at results, storage, can damage quality in transportation, comprises and damages, rots and germination etc.Damage, black change can occur the color of potato top layer, the position meat of rotting, if do not removed, the whiteness of potato powder can be reduced, causing the quality of product to reduce, decline in benefits.The potato meat at germination position can turn green, and this can reduce the quality of potato powder equally; In addition, these potatos are processed help powder after, because germination position is poisonous, consumer is edible can impair one's health.
For the rejecting of these potatos inferior, be adopt manual method now, examine the potato after decortication by naked eyes, find blackening, turn green potato, reject in time.This method wastes time and energy, and is based upon the recognition methods on people's sense organ, is subject to the behavioral implications of people, relevant with the visual cognitive ability degree of people, mood, degree of fatigue, thus can not identify these potatos inferior very well, accurately.
The basis of thermal imaging is the infrared radiation sent according to material.Infrared) radiation is the one electromagenetic wave radiation the most widely that nature exists, all are higher than the object of absolute zero (-273 DEG C), all constantly emitting infrared radiation, and simultaneously this radiation is all loaded with the characteristic information of object.Infrared thermal imaging technique utilizes object each several part temperature contrast to cause infrared emanation different, and sightless for object heat radiation situation is converted to visual thermography, thus the technology of the information of acquisition.If body surface or inner existing defects, then its surface temperature field is different, so also can there is difference to extraradial infrared energy, can reach the object of nondestructive testing on object quality accordingly.
Summary of the invention
The object of the present invention is to provide the defect inspection method of a kind of potato, to improve efficiency and the accuracy rate of the defects detection of potato.
In order to solve above technical matters, the present invention utilizes thermal imaging to reach the peeled potatoes identifying fast and damage, rot and germinate, and concrete technical scheme is as follows:
A defect inspection method for potato, is characterized in that comprising the following steps:
Step one, by the potato after steam peeling, with the water flow cleaning 3 ~ 5min of 5-30 DEG C, take out, unheated air dries up;
Step 2, transfers on sheet metal by step one gained potato, contacts the flowing water of 5-30 DEG C under plate, and keep temperature stabilization, heat potato with infrared lamp, bulb power 250 ~ 500W, the distance of bulb and potato is 35 ~ 55cm, heating duration 0.5 ~ 2min;
Step 3, takes pictures with thermal camera, and camera lens is apart from potato 35 ~ 55cm, and the infrared photograph of shooting is transferred to computing machine;
Step 4, computing machine potato thermal imagery carries out Treatment Analysis, identifies defective potato.
The thermal imagery processing procedure of potato thermal imagery being carried out to Treatment Analysis is: carry out medium filtering or mean filter denoising to thermal imagery, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area S of the corresponding material object of pixel of 1 DEG C, if S is greater than 25 mm2, is then defective potato.
The present invention has beneficial effect.The advantage that the present invention has and beneficial effect be mainly reflected in following some: the present invention is by machine recognition potato inferior, the existing shortcoming by the identification of people's sense organ can be overcome as the accuracy being subject to mood, mood impact identifies, thus improve identification normal solution rate, and good stability; The present invention is compared with artificial cognition, and machine recognition can reduce production cost greatly, puies forward the efficiency that original text identifies.
Embodiment
Below by specific embodiment, the present invention is described in further detail.
embodiment 1
1 by the potato after steam peeling 50, and with the water flow cleaning 15min of 5 DEG C, take out, unheated air dries up.
Potato is transferred on sheet metal by 2, contacts the flowing water of 15 DEG C, keep temperature stabilization under plate.Heat potato with infrared lamp, bulb power 250W, the distance of bulb and potato is 35cm, heating duration 0.5min.
3 take pictures with thermal camera, and camera lens is apart from potato 35cm, and the thermal imagery of shooting is transferred to computing machine.
4 pairs of thermal imagerys carry out mean filter denoising, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area of the corresponding material object of pixel of 1 DEG C
sif,
sbe greater than 25 mm
2, be then defective potato.
Through artificial nucleus couple, the method obtain thermal imagery machine recognition accuracy reach 92.5%.
embodiment 2
1 by the potato after steam peeling 50, and with the water flow cleaning 10min of 15 DEG C, take out, unheated air dries up.
Potato is transferred on sheet metal by 2, contacts the flowing water of 30 DEG C, keep temperature stabilization under plate.Heat potato with infrared lamp, bulb power 250W, the distance of bulb and potato is 80cm, heating duration 1min.
3 take pictures with thermal camera, and camera lens is apart from potato 80cm, and the infrared photograph of shooting is transferred to computing machine.
4 pairs of thermal imagerys carry out mean filter denoising, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area of the corresponding material object of pixel of 1 DEG C
sif,
sbe greater than 25 mm
2, be then defective potato.
Through artificial nucleus couple, the method obtain thermal imagery machine recognition accuracy reach 95%.
embodiment 3
1 by the potato after steam peeling 50, and with the water flow cleaning 3min of 30 DEG C, take out, unheated air dries up.
Potato is transferred on sheet metal by 2, contacts the flowing water of 5 DEG C, keep temperature stabilization under plate.Heat potato with infrared lamp, bulb power 250W, the distance of bulb and potato is 55cm, heating duration 40 s.
3 take pictures with thermal camera, and camera lens is apart from potato 55cm, and the infrared photograph of shooting is transferred to computing machine.
4 pairs of thermal imagerys carry out mean filter denoising, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area of the corresponding material object of pixel of 1 DEG C
sif,
sbe greater than 25 mm
2, be then defective potato.
Through artificial nucleus couple, the method obtain thermal imagery machine recognition accuracy reach 97.5%.
embodiment 4
1 by the potato after steam peeling 50, and with the water flow cleaning 4min of 20 DEG C, take out, unheated air dries up.
Potato is transferred on sheet metal by 2, contacts the flowing water of 15 DEG C, keep temperature stabilization under plate.Heat potato with infrared lamp, bulb power 500W, the distance of bulb and potato is 50cm, heating duration 0.25min.
3 take pictures with thermal camera, and camera lens is apart from potato 50cm, and the infrared photograph of shooting is transferred to computing machine.
4 pairs of thermal imagerys carry out mean filter denoising, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area of the corresponding material object of pixel of 1 DEG C
sif,
sbe greater than 25 mm
2, be then defective potato.
Through artificial nucleus couple, the method obtain thermal imagery machine recognition accuracy reach 95%.
embodiment 5
1 by the potato after steam peeling 50, and with the water flow cleaning 5min of 20 DEG C, take out, unheated air dries up.
Potato is transferred on sheet metal by 2, contacts the flowing water of 25 DEG C, keep temperature stabilization under plate.Heat potato with infrared lamp, bulb power 250W, the distance of bulb and potato is 50cm, heating duration 1min.
3 take pictures with thermal camera, and camera lens is apart from potato 50cm, and the infrared photograph of shooting is transferred to computing machine.
4 pairs of thermal imagerys carry out mean filter denoising, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area of the corresponding material object of pixel of 1 DEG C
sif,
sbe greater than 25 mm
2, be then defective potato.
Through artificial nucleus couple, the method obtain thermal imagery machine recognition accuracy reach 92.5%.
embodiment 6
1 by the potato after steam peeling 50, and with the water flow cleaning 8min of 15 DEG C, take out, unheated air dries up.
Potato is transferred on sheet metal by 2, contacts the flowing water of 300 DEG C, keep temperature stabilization under plate.Heat potato with infrared lamp, bulb power 100W, the distance of bulb and potato is 40cm, heating duration 2min.
3 take pictures with thermal camera, and camera lens is apart from potato 40cm, and the infrared photograph of shooting is transferred to computing machine.
4 pairs of thermal imagerys carry out mean filter denoising, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area of the corresponding material object of pixel of 1 DEG C
sif,
sbe greater than 25 mm
2, be then defective potato.
Through artificial nucleus couple, the method obtain thermal imagery machine recognition accuracy reach 90%.
Claims (2)
1. a defect inspection method for potato, is characterized in that comprising the following steps:
Step one, by the potato after steam peeling, with the water flow cleaning 3 ~ 5min of 5-30 DEG C, take out, unheated air dries up;
Step 2, transfers on sheet metal by step one gained potato, contacts the flowing water of 5-30 DEG C under plate, and keep temperature stabilization, heat potato with infrared lamp, bulb power 250 ~ 500W, the distance of bulb and potato is 35 ~ 55cm, heating duration 0.5 ~ 2min;
Step 3, takes pictures with thermal camera, and camera lens is apart from potato 35 ~ 55cm, and the infrared photograph of shooting is transferred to computing machine;
Step 4, computing machine carries out Treatment Analysis to potato thermal imagery, identifies defective potato.
2. the defect inspection method of a potato as claimed in claim 1, it is characterized in that describedly to the thermal imagery processing procedure that potato thermal imagery carries out Treatment Analysis be: medium filtering or mean filter denoising are carried out to thermal imagery, adopt threshold segmentation method, extract the region of potato thermal imagery, calculate the average temperature value of pixel in this region, then the temperature value of each pixel and the difference of average temperature value in zoning, the absolute value of calculated difference is greater than the area S of the corresponding material object of pixel of 1 DEG C, goes out defective potato according to this area recognition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310280939.3A CN103344647B (en) | 2013-07-05 | 2013-07-05 | The defect inspection method of a kind of potato |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310280939.3A CN103344647B (en) | 2013-07-05 | 2013-07-05 | The defect inspection method of a kind of potato |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103344647A CN103344647A (en) | 2013-10-09 |
| CN103344647B true CN103344647B (en) | 2015-10-28 |
Family
ID=49279461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310280939.3A Expired - Fee Related CN103344647B (en) | 2013-07-05 | 2013-07-05 | The defect inspection method of a kind of potato |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103344647B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106383114A (en) * | 2016-08-30 | 2017-02-08 | 安徽丰絮农业科技股份有限公司 | Potato seed quality detection method |
| CN106872520A (en) * | 2017-02-07 | 2017-06-20 | 贵阳学院 | Using the stealthy method damaged of infrared image temperature difference check and evaluation fruit surface |
| CN106872044A (en) * | 2017-02-07 | 2017-06-20 | 贵阳学院 | Method based on infrared image monitoring and warning crops early disease and infected zone |
| CN106841305A (en) * | 2017-02-07 | 2017-06-13 | 贵阳学院 | Based on the stealthy method damaged of infrared image Non-Destructive Testing fruit surface |
| CN106596566A (en) * | 2017-02-07 | 2017-04-26 | 贵阳学院 | Method for testing and comprehensively evaluating invisible damage to surface of fruit through infrared images |
| CN106872473A (en) * | 2017-02-21 | 2017-06-20 | 中国矿业大学 | A kind of potato defects detection identifying system design based on machine vision |
| CN107742132A (en) * | 2017-11-07 | 2018-02-27 | 江南大学 | Potato detection method of surface flaw based on convolutional neural networks |
| CN113295546A (en) * | 2021-05-10 | 2021-08-24 | 武汉精测电子集团股份有限公司 | FPC microcircuit bending damage degree verification method and device and electronic equipment |
-
2013
- 2013-07-05 CN CN201310280939.3A patent/CN103344647B/en not_active Expired - Fee Related
Non-Patent Citations (3)
| Title |
|---|
| Non-contact bruise detection in apples by thermal imaging;J. Varitha;《Innovative Food Science and Emerging Technologies》;20030630;第4卷(第2期);第211-218页 * |
| 基于主动热成像技术的苹果表面缺陷分类方法;周建民，张瑞丰;《华东交通大学学报》;20120229;第29卷(第1期);第86-89页 * |
| 苹果早期机械损伤的红外热成像检测研究;周其显;《中国优秀硕士学位论文全文数据库》;20120515;说明书第17-20页第3.1.2节，第38页第5.2节，第50页第5.4.1节，图3-3 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103344647A (en) | 2013-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103344647B (en) | The defect inspection method of a kind of potato | |
| CN103592316A (en) | Method for detecting defects of potatoes | |
| Ji et al. | Non-destructive classification of defective potatoes based on hyperspectral imaging and support vector machine | |
| Pu et al. | Prediction of moisture content uniformity of microwave-vacuum dried mangoes as affected by different shapes using NIR hyperspectral imaging | |
| Zhang et al. | Combined medium-and short-wave infrared and hot air impingement drying of sponge gourd (Luffa cylindrical) slices | |
| EP2471024B1 (en) | Method for real time detection of defects in a food product | |
| Jiang et al. | Effects of combined radio frequency with hot water blanching on enzyme inactivation, color and texture of sweet potato | |
| Kaveh et al. | Impacts of hybrid (convective‐infrared‐rotary drum) drying on the quality attributes of green pea | |
| CN107044967B (en) | Method for quickly identifying near infrared spectrum of raw whole potato flour | |
| CN107568652A (en) | A kind of processing method and water chestnut crisp chip of the instant water chestnut crisp chip of high-quality | |
| CN113256575B (en) | Fruit defect detection method based on structured hyperspectral system | |
| Hao et al. | Investigation of the data fusion of spectral and textural data from hyperspectral imaging for the near geographical origin discrimination of wolfberries using 2D-CNN algorithms | |
| Sharma et al. | Application of computer vision technique for physical quality monitoring of turmeric slices during direct solar drying | |
| CN109682821B (en) | Citrus surface defect detection method based on multi-scale Gaussian function | |
| CN105029282B (en) | Pure natural steaming potato food and preparation method thereof | |
| Zenoozian et al. | Image analysis and dynamic modeling of thin‐layer drying of osmotically dehydrated pumpkin | |
| CN108850129A (en) | A kind of freeze-drying Fruit quality system and its control method | |
| Przybył et al. | Health properties and evaluation of quality of dried strawberry fruit produced using the convective drying method with neural image analysis | |
| Obajemihi et al. | Novel sequential and simultaneous infrared-accelerated drying technologies for the food industry: Principles, applications and challenges | |
| Hafezi et al. | The effect of drying kinetic on shrinkage and colour of potato slices in the vacuum-infrared drying method | |
| CN101543277B (en) | Expansion method for shaping jujubes | |
| Han et al. | Classification of glycyrrhiza seeds by near infrared hyperspectral imaging technology | |
| CN106596566A (en) | Method for testing and comprehensively evaluating invisible damage to surface of fruit through infrared images | |
| CN106173599A (en) | The Rhizoma Dioscoreae powder processing technology that a kind of brew is good | |
| CN105192618A (en) | Tempering and drying technology for Chinese yam slices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151028 Termination date: 20170705 |