TW202317936A - Method for estimating grain moisture content and harvest time of grains with mobile phone images - Google Patents
Method for estimating grain moisture content and harvest time of grains with mobile phone images Download PDFInfo
- Publication number
- TW202317936A TW202317936A TW110138643A TW110138643A TW202317936A TW 202317936 A TW202317936 A TW 202317936A TW 110138643 A TW110138643 A TW 110138643A TW 110138643 A TW110138643 A TW 110138643A TW 202317936 A TW202317936 A TW 202317936A
- Authority
- TW
- Taiwan
- Prior art keywords
- mobile phone
- image
- moisture content
- grain
- ear
- Prior art date
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
Description
本發明是關於一種一種通過萃取穀粒色彩特徵推估穀粒含水量及收穫時期的方法。The invention relates to a method for estimating grain water content and harvest time by extracting grain color features.
許多穀粒在不同的生長時期會有特定的顏色變化,以水稻為例,水稻的稻穗成熟時,會歷經乳熟、糊熟、黃熟與完熟等階段,部分農民是根據經驗來決定是否可以採收,然此判斷方式難以量化,唯一的科學根據為穀粒含水量,其數值不僅攸關收購方的收購價格,也影響烘乾所需的能源及稻米品質等。Many grains have specific color changes in different growth periods. Taking rice as an example, when rice ears mature, they will go through stages such as milk ripening, paste ripening, yellow ripening and full ripening. Some farmers decide whether to It can be harvested, but it is difficult to quantify the judgment method. The only scientific basis is the moisture content of the grain. The value not only affects the purchase price of the purchaser, but also affects the energy required for drying and the quality of rice.
穀粒含水量的評估方式除了根據經驗之外,也能透過電阻式水分計或烘乾量測含水量的方式來評估,前者容易受環境濕度影響準確性,後者單次量測耗時一週以上,無法即時得到量測結果,並且,兩種評估方式都涉及破壞性採樣,需將樣本從植株上取下。In addition to experience, the evaluation method of grain moisture content can also be evaluated through a resistance moisture meter or a drying method to measure the moisture content. The former is easily affected by the accuracy of the ambient humidity, and the latter takes more than a week for a single measurement , the measurement results cannot be obtained immediately, and both evaluation methods involve destructive sampling, and the samples need to be removed from the plants.
另一方面,根據農會的契作農民繳交濕穀統計資料,發現平均收獲穀粒含水量約在31%左右,而非建議較適合採收的26%,依照濕穀收購價格來看,若全台灣採收濕穀每下降1%,就可提升稻農每年11億的總體收益,也可以減少2.6億的烘乾燃油成本,與30萬公噸的碳排放量。因此,如何提供一種可以即時取得評估結果、並且不需破壞樣本的含水量評估方法,實是值得思量的。On the other hand, according to the statistic data of wet grain submitted by contracted farmers of the farmers association, it is found that the average harvested grain moisture content is about 31%, rather than 26% which is recommended to be more suitable for harvesting. According to the purchase price of wet grain, If the harvesting of wet rice in Taiwan is reduced by 1%, the overall annual income of rice farmers can be increased by 1.1 billion, and the cost of drying fuel by 260 million can also be reduced, as well as the carbon emissions of 300,000 metric tons. Therefore, it is worth considering how to provide a water content evaluation method that can obtain the evaluation results in real time and does not need to destroy the sample.
有鑑於此,本發明的其中一個目的是提供一種可以即時取得評估結果、並且不需破壞樣本的穀粒含水量評估方法。In view of this, one of the objectives of the present invention is to provide a method for evaluating grain moisture content that can obtain evaluation results immediately without destroying samples.
為了實現上述及其他目的,本發明提供一種以手機影像推估穀粒含水量及收穫時期的方法,其涉及使用一輔助拍攝工具,輔助拍攝工具包括一底架、一手機固定架、一攝影背板、至少一用以在該攝影背板標記一圖像輪廓的定位圖標以及至少一校正色板,手機固定架及攝影背板均設於底架且彼此間隔一段距離,定位圖標及校正色板均設於攝影背板,且圖像輪廓、定位圖標及校正色板均位於手機攝像鏡頭的拍攝範圍內;In order to achieve the above and other purposes, the present invention provides a method for estimating grain moisture content and harvest time using mobile phone images. Board, at least one positioning icon used to mark an image outline on the photographic backplane, and at least one calibration color board, the mobile phone fixing frame and the photographic backplane are both arranged on the bottom frame and spaced from each other at a certain distance, the positioning icon and the calibration color board They are all set on the back panel of the camera, and the image outline, positioning icons and calibration color plates are all located within the shooting range of the camera lens of the mobile phone;
其中,該方法包括: 以該手機固定架固定所述手機,將一穀穗放置於該攝影背板的所述圖像輪廓內,並以該攝像鏡頭拍攝一穀穗影像; 對該穀穗影像進行預處理,將該穀穗中的穀粒以外的背景雜訊濾除; 萃取預處理後的所述穀穗影像中的顏色特徵;以及 根據該顏色特徵估算所述穀粒的的含水率,並根據所估算的含水率及該穀穗影像的拍攝時間提供一建議採收日期。 Among them, the method includes: fixing the mobile phone with the mobile phone fixing frame, placing an ear of grain in the outline of the image on the photographic backboard, and taking an image of an ear of grain with the camera lens; Preprocessing the ear image to filter out background noise other than grains in the ear; extracting the color features in the preprocessed ear image; and Estimate the moisture content of the grain according to the color feature, and provide a suggested harvest date according to the estimated moisture content and the shooting time of the ear image.
基此,本發明可讓稻農透過手機作為穀粒含水量測量之工具,解決難以精準掌握穀粒含水量而使收益受損或收獲排程延宕之問題,提升稻農收益、稻米收穫效率,並減少烘乾稻穀所需能源與碳排量。Based on this, the present invention allows rice farmers to use mobile phones as a tool for measuring grain moisture content, solving the problem that it is difficult to accurately grasp the moisture content of grains, resulting in loss of income or delay in harvesting schedules, and improving rice farmers' income and rice harvesting efficiency. And reduce the energy and carbon emissions required for drying rice.
請參考第1至3圖,所繪示者為本發明使用的輔助拍攝工具的實施例,其包括一底架10、一手機固定架20、一攝影背板30、至少一定位圖標40及至少一校正色板50。Please refer to Figs. 1 to 3, which are the embodiments of the auxiliary shooting tool used in the present invention, which include a
底架10包括一握把11及一連接於握把11頂部的支撐桿12,手機固定架20及攝影背板30均設於底架10,分別位於支撐桿12的兩側,手機固定架20及攝影背板30保持固定的間隔以固定拍攝條件。手機固定架20是用以固定手機,手機上配置有攝像鏡頭,可用於拍攝樣品影像。為了便於收納,本實施例的手機固定架20及攝影背板30可相對底架10旋轉收折(如第3圖),減少佔用空間。
攝影背板30的底色為黑色,定位圖標40及校正色板50設於攝影背板30上。定位圖標40是用以在攝影背板30標記一圖像輪廓,本實施例中定位圖標40是位於圖像輪廓四個角落的白色三角形圖塊,以便於影像處理識別。校正色板50是用來作為色彩校正的參考,本實施例中包含了黑色、灰色及白色的校正色板,後續影像處理時,可減少不同環境光源對色彩萃取的影響。圖像輪廓、定位圖標40及校正色板50都位於手機攝像鏡頭的拍攝範圍內,因此手機拍攝影像時,至少會包括以上三個部分。The background color of the
以下說明本發明的工作流程:The workflow of the present invention is described below:
首先,以手機固定架固定手機,將預評估含水率的穀穗(例如稻穗)放置於攝影背板的前述圖像輪廓內,並以攝像鏡頭拍攝一穀穗影像(如第4圖所示),此時穀穗不需要從植株上移除,可在直接田間進行拍攝,亦即,本發明可以進行非破化性採樣,並且,可以針對田間的不同植株進行採樣,以降低採樣的個別誤差;First, fix the mobile phone with the mobile phone holder, place the grain ears (such as rice ears) with pre-evaluated moisture content in the aforementioned image outline on the photography backplane, and take a grain ear image with the camera lens (as shown in Figure 4 ), at this time, the ear does not need to be removed from the plant, and it can be photographed directly in the field, that is, the present invention can perform nondestructive sampling, and can sample different plants in the field to reduce the individual error;
接著,通過運算程式對穀穗影像進行預處理,將穀穗中的穀粒以外的背景雜訊濾除,所述預處理包括一系列的處理流程,例如,對影像進行二元化處理、通過定位圖標的標定移除圖像輪廓以外其他區域的影像、將定位圖標自穀穗影像中移除、根據灰階值移除穀穗影像中的陰影部分、根據顏色特徵移除影像中非穀粒的部分(例如穀穗中的莖),如此,穀穗影像中只會留下穀粒的部分(如第5圖所示);Next, the grain ear image is preprocessed by an algorithm to filter out background noise other than the grains in the grain ear. The preprocessing includes a series of processing procedures, for example, binary processing of the image, through The calibration of the positioning icon removes images in areas other than the outline of the image, removes the positioning icon from the ear image, removes the shadow part of the ear image according to the gray scale value, and removes the non-grain in the image according to the color feature The part of the ear (such as the stem in the ear), so that only the part of the grain will be left in the ear image (as shown in Figure 5);
再接著,即可通過運算程式將預處理後的穀穗影像中的顏色特徵萃取出來;Then, the color features in the preprocessed grain ear image can be extracted through the calculation program;
最後,運算程式可根據所萃取的顏色特徵估算穀粒的含水率,並根據所估算的含水率及穀穗影像的拍攝時間提供建議採收日期;其中,含水率的估算可通過機器學習的演算法,利用大量的穀穗影像樣本由演算法建立含水率估算模型,後續即可利用已建立的含水率估算模型推估穀粒含水率,建議採收日期的部分,穀粒在熟成時含水率會逐漸下降,其下降率可建立一回歸公式,因此運算程序可根據該回歸公式推估所採樣的穀穗樣本的穀粒含水率降低至目標含水率的所需時間,從而推估出建議採收日期。Finally, the calculation program can estimate the moisture content of the grains based on the extracted color features, and provide a suggested harvest date based on the estimated moisture content and the shooting time of the ear image; the moisture content can be estimated through machine learning calculations method, using a large number of ear image samples to establish a moisture content estimation model by algorithm, and then use the established moisture content estimation model to estimate the moisture content of grains. It is recommended that the part of the harvest date, the moisture content of the grains when they mature It will gradually decrease, and its rate of decline can establish a regression formula, so the calculation program can estimate the time required for the grain moisture content of the sampled grain ear samples to decrease to the target moisture content according to the regression formula, so as to estimate the recommended harvesting time. receipt date.
其中,上述運算的執行不需要全部由手機內建的處理器執行,一部份運算工作(例如預處理及顏色特徵的萃取)可由一雲端伺服器來執行,手機只要至少提供操作界面、上傳穀穗影像至雲端伺服器並顯示結果即可。Among them, the execution of the above calculations does not need to be all performed by the built-in processor of the mobile phone. Part of the calculation work (such as preprocessing and color feature extraction) can be performed by a cloud server. The spike image is sent to the cloud server and the result is displayed.
10:底架 11:握把 12:支撐桿 20:手機固定架 30:攝影背板 40:定位圖標 50:校正色板 10: Chassis 11: Grip 12: Support rod 20:Mobile phone holder 30: Photography backplane 40: Positioning icon 50: Correction Swatches
第1圖是本發明其中一實施例的使用狀態立體圖。Figure 1 is a perspective view of one embodiment of the present invention in use.
第2圖是本發明其中一實施例的攝影背板的前視圖。Fig. 2 is a front view of a photographic backplane according to one embodiment of the present invention.
第3圖是本發明其中一實施例的折疊示意圖。Fig. 3 is a schematic folding diagram of one embodiment of the present invention.
第4、5圖是穀穗影像的示意圖。Figures 4 and 5 are schematic diagrams of grain ear images.
10:底架 10: Chassis
11:握把 11: Grip
12:支撐桿 12: Support rod
20:手機固定架 20:Mobile phone holder
30:攝影背板 30: Photography backplane
40:定位圖標 40: Positioning icon
50:校正色板 50: Correction Swatches
Claims (3)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW110138643A TWI818334B (en) | 2021-10-19 | 2021-10-19 | Method for estimating grain moisture content and harvest time of grains with mobile phone images |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW110138643A TWI818334B (en) | 2021-10-19 | 2021-10-19 | Method for estimating grain moisture content and harvest time of grains with mobile phone images |
Publications (2)
Publication Number | Publication Date |
---|---|
TW202317936A true TW202317936A (en) | 2023-05-01 |
TWI818334B TWI818334B (en) | 2023-10-11 |
Family
ID=87378790
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW110138643A TWI818334B (en) | 2021-10-19 | 2021-10-19 | Method for estimating grain moisture content and harvest time of grains with mobile phone images |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI818334B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117409403A (en) * | 2023-12-15 | 2024-01-16 | 南京农业大学三亚研究院 | Rice spike maturity estimation method based on deep learning |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101328487B1 (en) * | 2011-12-27 | 2013-11-13 | 한국식품연구원 | Prediction Method and Device of Rough Rice Harvesting Time |
US10157472B2 (en) * | 2015-03-12 | 2018-12-18 | Raytheon Company | Apparatus and processes for corn moisture analysis and prediction of optimum harvest date |
US11010913B2 (en) * | 2016-10-19 | 2021-05-18 | Basf Agro Trademarks Gmbh | Determining the grain weight of an ear |
JP6267841B1 (en) * | 2017-04-28 | 2018-01-24 | 株式会社オプティム | Wearable terminal display system, wearable terminal display method and program |
CN108387534A (en) * | 2018-03-15 | 2018-08-10 | 中国农业科学院农田灌溉研究所 | Measure plant water content method and apparatus |
CN111289441B (en) * | 2020-02-21 | 2021-02-26 | 中国农业大学 | Multispectral field crop water content determination method, system and equipment |
CN111948734A (en) * | 2020-06-29 | 2020-11-17 | 北京农业信息技术研究中心 | Crop canopy phenotype and microclimate parameter measuring device and method |
CN212624062U (en) * | 2020-08-24 | 2021-02-26 | 甘肃省农业科学院小麦研究所 | Auxiliary device for image acquisition |
-
2021
- 2021-10-19 TW TW110138643A patent/TWI818334B/en active
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117409403A (en) * | 2023-12-15 | 2024-01-16 | 南京农业大学三亚研究院 | Rice spike maturity estimation method based on deep learning |
CN117409403B (en) * | 2023-12-15 | 2024-03-19 | 南京农业大学三亚研究院 | Rice spike maturity estimation method based on deep learning |
Also Published As
Publication number | Publication date |
---|---|
TWI818334B (en) | 2023-10-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Vesali et al. | Development of an android app to estimate chlorophyll content of corn leaves based on contact imaging | |
Makanza et al. | High-throughput method for ear phenotyping and kernel weight estimation in maize using ear digital imaging | |
CN102506772B (en) | Method and device for quickly detecting area of leaf blade based on mobile phone | |
TW202317936A (en) | Method for estimating grain moisture content and harvest time of grains with mobile phone images | |
CN111462058B (en) | Method for rapidly detecting effective rice ears | |
CN112345458A (en) | Wheat yield estimation method based on multispectral image of unmanned aerial vehicle | |
Roth et al. | Repeated multiview imaging for estimating seedling tiller counts of wheat genotypes using drones | |
Cai et al. | Quantifying the onset and progression of plant senescence by color image analysis for high throughput applications | |
CN111797760A (en) | Improved crop pest and disease identification method based on Retianet | |
Constantino et al. | Plant height measurement and tiller segmentation of rice crops using image processing | |
CN113850785A (en) | Method and system for generating and evaluating image quality model, and equipment and medium thereof | |
CN106023235A (en) | Crop effective grain number measuring method | |
JP6721218B2 (en) | Grapes counting method | |
CN112098415B (en) | Nondestructive testing method for quality of waxberries | |
US8655102B2 (en) | Method and system for identifying tokens in an image | |
WO2020179276A1 (en) | Image processing device, image processing method, and program | |
CN102239793A (en) | Real-time classification method and system of rice pests | |
JP2019138831A (en) | Color index value acquisition system, and color index value acquisition method | |
CN116308435A (en) | Agricultural product precise identification method and system based on intelligent traceability scale | |
CN115687850A (en) | Method and device for calculating irrigation water demand of farmland | |
CN113506343B (en) | Color coordinate estimation method, system, device and medium based on multi-source data | |
CN110765905A (en) | Method and device for measuring specific gravity of impurities contained in grains harvested by combine harvester | |
Zhang et al. | Research on a rapid identification method for counting universal grain crops | |
CN107131851A (en) | A kind of method for rapidly and efficiently measuring sample prescription implants cover degree | |
TWI838787B (en) | Rice leaf nitrogen content detection method and system thereof using rice leaf images |