WO2020233156A1 - 一种大姜精量定向种植中鳞芽识别及调整方法 - Google Patents

一种大姜精量定向种植中鳞芽识别及调整方法 Download PDF

Info

Publication number
WO2020233156A1
WO2020233156A1 PCT/CN2020/072126 CN2020072126W WO2020233156A1 WO 2020233156 A1 WO2020233156 A1 WO 2020233156A1 CN 2020072126 W CN2020072126 W CN 2020072126W WO 2020233156 A1 WO2020233156 A1 WO 2020233156A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
ginger
stage
scale
bud
Prior art date
Application number
PCT/CN2020/072126
Other languages
English (en)
French (fr)
Inventor
杨发展
曹铭恺
王树成
王超
王鑫
杜祥汶
李维华
Original Assignee
青岛理工大学
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 青岛理工大学 filed Critical 青岛理工大学
Publication of WO2020233156A1 publication Critical patent/WO2020233156A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01C—PLANTING; SOWING; FERTILISING
    • A01C1/00—Apparatus, or methods of use thereof, for testing or treating seed, roots, or the like, prior to sowing or planting
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D3/00—Control of position or direction
    • G05D3/10—Control of position or direction without using feedback
    • 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
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G06V10/242—Aligning, centring, orientation detection or correction of the image by image rotation, e.g. by 90 degrees

Definitions

  • the invention relates to a method for identifying and adjusting scale buds in precise directional planting of ginger, and belongs to the technical field of image recognition.
  • Ginger also known as ginger, has a spicy flavor and belongs to the genus Zingiberaceae. It is a perennial herbaceous perennial herb. It is cultivated as an annual economic crop in my country and is an important vegetable variety that is a specialty of my country. Ginger is generally used as a food material and is an economical crop. Ginger is cultivated as an annual cash crop in my country, and the level of mechanized planting of Ginger is low. The key factor for ginger planting is the field placement of ginger scale buds so that the scale buds face south to have a suitable growth environment.
  • the ginger precision directional planter can help improve the planting efficiency of ginger, reduce the damage rate of ginger seeds and save labor costs, which has important practical significance.
  • the scale bud recognition strategy based on image processing is an important key technology for ginger planting machinery to realize ginger planting.
  • the identification and adjustment of scale buds is of great significance to the quality of ginger species, to ensure that the scale buds are facing the sun and have the most suitable growth environment.
  • the purpose of the present invention is to overcome the shortcomings of the prior art, and propose a method for precise directional planting and adjustment of the direction of ginger scale buds based on image recognition.
  • a method for identifying and adjusting scale buds in precise directional planting of ginger including HSV transformation stage, color segmentation stage, image binarization stage, corrosion operation stage, expansion operation stage, image recognition stage, deviation calculation stage and scale bud adjustment stage, Specific steps are as follows:
  • the first step HSV conversion stage, collect the image of the HSV color model of the ginger species or convert the image of other color models of the ginger species to the HSV color model and store it in the memory;
  • Step 2 In the color segmentation stage, perform color segmentation on the image, and retain the color area of ginger scale buds in the image;
  • the third step the image binarization stage, the retained image is binarized
  • Step 4 Corrosion operation is performed on the binary image
  • Step 5 In the expansion operation stage, perform expansion operation on the corroded binary image
  • the sixth step the image recognition stage, identify the position of ginger scale buds in the image
  • Step 7 In the deviation calculation stage, calculate the angular deviation of ginger scale buds relative to the 0 degree baseline;
  • Step 8 In the stage of scale bud adjustment, the stepper motor adjustment device is controlled to rotate with the ginger ginger seeds during planting to realize the adjustment of the direction of the ginger scale buds.
  • the color segmentation stage refers to the color segmentation operation on the image of the HSV color model of the ginger species by setting the characteristic parameters of hue H, saturation S, and lightness V, and only the colors in [26, 43 , 120] and [77, 200, 255].
  • the image binarization stage refers to performing a binarization operation on the part of the image left in the color segmentation stage to convert the image into a black and white image.
  • the erosion operation stage refers to performing an erosion operation on the image obtained after the image binarization stage to effectively filter out the noise in the image after the binarization process without destroying the original information of the image, and extract the smooth the edge of.
  • the expansion operation stage refers to performing expansion processing on the image after the corrosion operation stage, filling the image boundary to smooth the image boundary, making each area of the image more distinct, and at the same time not significantly changing the original size of the image area.
  • the image recognition stage refers to finding independent color regions in the image obtained after the expansion operation stage, and calculating the size and position of each region, so as to obtain the size and position of each ginger scale bud.
  • the deviation calculation stage refers to calculating the position A of a new ginger scale bud using a weighted average algorithm, and then calculating the angular deviation between the position A and the 0 degree baseline.
  • the scale bud adjustment stage refers to the reference position set before planting, and the stepping motor adjustment device is controlled to drive the position A of the new ginger scale bud to rotate to an angle coincident with the reference position.
  • the stepping motor adjusting device includes a stepping motor, a gear set arranged below the stepping motor, and a rotating shaft set below the gear set;
  • the gear set includes a first gear and The second gear;
  • the first gear is coaxially fixed with the rotating shaft of the stepping motor;
  • the second gear is coaxially fixed with the rotating shaft;
  • under the rotating shaft is also placed ginger with ginger scales on the surface Ginger species;
  • the upper end of the pin is fixed to the rotating shaft, and the lower end of the pin is inserted into the ginger species;
  • the stepping motor drives the first gear to rotate, and the second gear can drive the rotating shaft to rotate, thereby driving the big Adjustment of the direction of ginger scale buds on ginger ginger plants.
  • the expansion operation after the corrosion operation can maximize the integrity of the subsequent scale bud images.
  • Fig. 1 is a block diagram of the scale bud identification and adjustment method of the present invention.
  • Fig. 2 is an original image of Zingiber officinale in the embodiment of the present invention.
  • Fig. 3 is a schematic diagram of image binarization output of the present invention.
  • Figure 4 is a schematic diagram of the output of the corrosion expansion operation of the present invention.
  • Figure 5 is a schematic diagram of the final position of the scale buds of the present invention.
  • Fig. 6 is a schematic structural diagram of the stepping motor adjusting device of the present invention.
  • the first step HSV conversion stage, collect the image of the HSV color model of Zingiber officinale 4 or convert the image of other color models of Zingiber officinale 4 to HSV color model and store it in the memory; as shown in Figure 2 Original image of ginger ginger species 4;
  • Step 2 In the color segmentation stage, perform color segmentation operations on the image, and retain the 6 color areas of ginger scale buds in the image;
  • the third step the image binarization stage, the retained image is binarized; as shown in Figure 3;
  • Step 4 Corrosion operation is performed on the binary image
  • Step 5 In the expansion operation stage, perform expansion operation on the corroded binary image; as shown in Figure 4;
  • Step 6 In the image recognition stage, identify the position of ginger scale bud 6 in the image;
  • Step 7 In the deviation calculation stage, calculate the angular deviation of ginger scale bud 6 relative to the 0 degree baseline;
  • Step 8 In the stage of adjusting scale buds, control the stepper motor adjustment device to rotate with ginger ginger species 4 during planting to realize the adjustment of ginger scale buds in 6 directions.
  • the color parameters in the HSV color model are: hue H, saturation S, and lightness V.
  • the hue H is measured by an angle, and the value range is 0° ⁇ 360°. It is calculated in a counterclockwise direction from red. Red is 0°, green is 120°, and blue is 240°. Their complementary colors are: 60° for yellow, 180° for cyan, and 300° for magenta.
  • Saturation S indicates how close the color is to the spectral color.
  • a color can be seen as the result of mixing a certain spectral color with white. Among them, the greater the proportion of the spectral color, the higher the degree of the color close to the spectral color, and the higher the saturation of the color. With high saturation, the color is deep and brilliant. The white light component of the spectral color is 0, and the saturation reaches the highest. Usually the value ranges from 0% to 100%. The larger the value, the more saturated the color.
  • Lightness V indicates the brightness of the color.
  • the lightness value is related to the brightness of the luminous body; for the object color, this value is related to the transmittance or reflectance of the object. Usually the value ranges from 0% (black) to 100% (white).
  • the color segmentation stage refers to the color segmentation operation on the image of the HSV color model of Zingiber officinale 4 by setting the characteristic parameters of hue H, saturation S, and lightness V, and only the colors in [26, The pixels between 43, 120] and [77, 200, 255].
  • the hue H, saturation S, and lightness V in the above two matrix ranges can completely include the color of general ginger scale buds 6, so the position of ginger buds can be identified very quickly by this method.
  • the image binarization stage refers to performing a binarization operation on the part of the image left in the color segmentation stage to convert the image into a black and white image.
  • the entire image presents a clear black and white effect, and the binarization of the image greatly reduces the amount of data in the image, which can highlight the outline of the target.
  • the threshold for binarization is calculated by the OTSU method (also known as the maximum between-class variance method).
  • the central idea of OTSU is that the threshold T should maximize the between-class variance between the target and the background.
  • the segmentation threshold between the current scene and the background is t
  • the proportion of the front scenic spot in the image is w 0
  • the mean is u 0
  • the proportion of background points in the image is w 1
  • the mean is u 1
  • the objective function g(t) w 0 ⁇ (u 0 -u) 2 +w 1 ⁇ (u 1 -u) 2
  • g(t) is the between-class variance expression when the segmentation threshold is t .
  • the OTSU algorithm makes g(t) obtain the global maximum value. When g(t) is the maximum, the corresponding t is called the optimal threshold.
  • the erosion operation stage refers to performing an erosion operation on the image obtained after the image binarization stage to effectively filter out the noise in the image after the binarization process without destroying the original information of the image, and extract the smooth the edge of.
  • the borders of the image after binarization are not smooth, the object area has some noise holes, and the background area is scattered with some small noise objects. Therefore, the corrosion operation on the binary image can effectively filter out the noise in the image obtained by the above steps without destroying the original information of the image, and the edges extracted by this algorithm are also relatively smooth.
  • the expansion operation stage refers to performing expansion processing on the image after the corrosion operation stage, and filling the image boundary makes the image boundary smoother and makes each area of the image more distinct without significantly changing the original size of the image area.
  • the image recognition stage refers to finding independent color regions in the image obtained after the expansion operation stage, and calculating the size and position of each region, so as to obtain the region size and position of each large ginger scale bud 6 (as shown in Figure 4).
  • the white area is the area of each ginger scale bud 6).
  • the deviation calculation stage refers to calculating the position A of a new ginger scale bud 6 using a weighted average algorithm (as shown in FIG. 5, the circle in the box on the left side of the thick black line is the position A). Then take the center of the image as the origin, and the 0-degree baseline of the origin as the X axis, draw a line from the center of the image to the center of position A, then the line (shown as the thick black line in Figure 5) and the 0-degree baseline The angle is the angular deviation between the position A of the new ginger scale bud 6 and the 0 degree baseline.
  • the scale bud adjustment stage refers to the reference position set before planting, and the stepping motor adjustment device is controlled to drive the position A of the new ginger scale bud 6 to rotate to an angle coincident with the reference position.
  • the stepping motor adjusting device includes a stepping motor 1, a gear set 2 arranged below the stepping motor 1, and a rotating shaft 3 arranged below the gear set 2;
  • the gear set 2 includes mutual Meshing first gear and second gear;
  • the first gear is coaxially fixed with the rotating shaft of the stepping motor 1;
  • the second gear is coaxially fixed with the rotating shaft 3; there is also under the rotating shaft 3
  • a ginger ginger species 4 with large ginger scale buds 6 on the surface is placed;
  • the upper end of a pin 5 is fixed to the rotating shaft 3, and the lower end of the pin 5 is inserted into the ginger ginger species 4;
  • the stepping motor 1 drives the first gear Rotation, the second gear can drive the rotating shaft 3 to rotate, so as to drive the 6-direction adjustment of the ginger scale buds on the ginger ginger seed 4.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Soil Sciences (AREA)
  • Environmental Sciences (AREA)
  • Automation & Control Theory (AREA)
  • Image Analysis (AREA)
  • Color Image Communication Systems (AREA)

Abstract

本发明涉及一种大姜精量定向种植中鳞芽识别及调整方法，包括HSV转化阶段：采集大姜姜种的HSV颜色模型的图像或将大姜姜种的其他彩色模型的图像转换为HSV颜色模型，并存储到内存中；颜色分割阶段：对图像进行颜色分割操作，保留下图像中的大姜鳞芽颜色区域；图像二值化阶段：对保留下来的图像进行二值化；腐蚀操作阶段：对二值化图像进行腐蚀操作；膨胀操作阶段：对腐蚀后的二值化图像进行膨胀操作；图像识别阶段：识别图像中大姜鳞芽位置；偏差计算阶段：计算大姜鳞芽相对于0度基线的角度偏差以及鳞芽调整阶段：在种植时控制步进电机调整装置带着大姜姜种旋转，实现大姜鳞芽方向的调整。

Description

一种大姜精量定向种植中鳞芽识别及调整方法 技术领域
本发明涉及一种大姜精量定向种植中鳞芽识别及调整方法，属于图像识别技术领域。
背景技术
大姜又称生姜，本味辣，属姜科姜属，为多年生草本宿根草本植物，在我国作为一年生经济作物栽培，是我国特产的重要蔬菜品种。大姜一般作为食材多用，是一种经济型农作物。大姜在我国作为一年生经济作物栽培，大姜的机械化种植水平低。而大姜种植的关键因素是对姜种鳞芽的实地摆放，使鳞芽朝南，以具备适宜的生长环境。
大姜精量定向种植机有助于提高大姜种植效率、降低姜种损伤率及节省人工成本，具有重要现实意义。其中基于图像处理的鳞芽识别策略是大姜种植机械实现大姜种植的重要关键技术。另一方面，鳞芽的识别调整对于姜种摆放质量有重要的意义，以保证鳞芽朝阳，具备最适宜的生长环境。尽管大姜种植中姜种的处理难度较大，特别是国内学者研究很少，但是其已逐渐成为国内外农业工程领域的一个新的研究热点。
目前，国内关于大姜鳞芽识别的相关专利暂时没有。经对现有文献检索发现类似研究，高迟等人在2010发表了题为《大蒜鳞芽方向识别的实验研究》的文章，该文章公开了一种大蒜鳞芽方向识别识别方法，设计一种特殊的斗形识别器。下部只能使蒜瓣的鳞芽穿过，当大蒜鳞芽朝下时由光电传感测得数据，由此判断大蒜鳞芽的方向。但由于大姜鳞芽的脆弱性以及姜种形状的不规则性，此方法无法应用于大姜鳞芽识别。
发明内容
本发明的目的在于克服现有技术的不足，提出了一种基于图像识别的精量定向种植和大姜鳞芽方向调整的方法。
本发明的技术方案如下：
一种大姜精量定向种植中鳞芽识别及调整方法，包括HSV转化阶段、颜色分割阶段、图像二值化阶段、腐蚀操作阶段、膨胀操作阶段、图像识别阶段、偏差计算阶段以及鳞芽调整阶段，具体步骤如下：
第一步：HSV转化阶段，采集大姜姜种的HSV颜色模型的图像或将大姜姜种的其他彩色模型的图像转换为HSV颜色模型，并存储到内存中；
第二步：颜色分割阶段，对图像进行颜色分割操作，保留下图像中的大姜鳞芽颜色区域；
第三步：图像二值化阶段，对保留下来的图像进行二值化；
第四步：腐蚀操作阶段，对二值化图像进行腐蚀操作；
第五步：膨胀操作阶段，对腐蚀后的二值化图像进行膨胀操作；
第六步：图像识别阶段，识别图像中大姜鳞芽位置；
第七步：偏差计算阶段，计算大姜鳞芽相对于0度基线的角度偏差；
第八步：鳞芽调整阶段，在种植时控制步进电机调整装置带着大姜姜种旋转，实现大姜鳞芽方向的调整。
进一步的，所述颜色分割阶段是指通过设定色调H、饱和度S、明度V的特征参数，对大姜姜种的HSV颜色模型的图像进行颜色分割操作，只保留颜色在[26，43，120]和[77，200，255]之间的像素点。
进一步的，所述图像二值化阶段是指将颜色分割阶段留下的图像部分进行二值化操作，将图像转换为黑和白的图像。
进一步的，所述腐蚀操作阶段是指对图像二值化阶段后得到的图像进行腐蚀操作，将二值化处理后图像中的噪点有效滤除而不破坏图像的原有信息，并提取出平滑的边缘。
进一步的，所述膨胀操作阶段是指对腐蚀操作阶段后的图像进行膨胀处理，填充图像边界使图像边界变得平滑，让图像各区域更分明，同时不会明显改变图像区域原来的大小。
进一步的，所述图像识别阶段是指在膨胀操作阶段后得到的图像中寻找独立的颜色区域，计算各个区域的大小和位置，从而得到每片大姜鳞芽的区域大小和位置。
进一步的，所述偏差计算阶段是指用加权平均算法计算出一个新的大姜鳞芽的位置A，再计算位置A与0度基线的角度偏差。
进一步的，所述鳞芽调整阶段是指在种植之前设定的基准位置，控制步进电机调整装置带动新的大姜鳞芽的位置A旋转到与基准位置角度重合。
进一步的，所述步进电机调整装置包括步进电机、设置于所述步进电机下方的齿轮组以及设置于所述齿轮组下方的旋转轴；所述齿轮组包括相互啮合的第一齿轮和第二齿轮；所述第一齿轮与所述步进电机的转轴同轴固定；所述第二齿轮与所述旋转轴同轴固定；所述旋转轴下方还放置有表面有大姜鳞芽的大姜姜种；有插针上端与所述旋转轴固定，所述插针下端插入所述大姜姜种；步进电机驱动第一齿轮转动，通过第二齿轮能够带动旋转轴转动，从而实现带动大姜姜种上大姜鳞芽方向的调整。
本发明具有如下有益效果：
1、本方法操作简单，过程简介明了，实施起来较为容易，同时精确度高。
2、使用本方法能够实现大姜鳞芽识别的自动化，识别速度快。
3、使用本方法识别大姜鳞芽，不会伤及鳞芽，能够对较为脆弱的鳞芽起到保护作用。
4、使用本方法，能够提高识别精度，减小因大姜姜种形状的不规则而引起的测量误差以及测量困难。
5、通过腐蚀操作，能够有效消除二值化图像中多余的噪点。
6、腐蚀操作后再进行膨胀操作，能够最大限度地保证后续鳞芽图像的完整性。
附图说明
图1为本发明的鳞芽识别及调整方法的框图。
图2为本发明实施例中大姜姜种的原始图像。
图3为本发明的图像二值化输出示意图。
图4为本发明的腐蚀膨胀操作输出示意图。
图5为本发明的鳞芽最终位置判别示意图。
图6为本发明的步进电机调整装置的结构示意图。
图中附图标记表示为：
1、步进电机；2、齿轮组；3、旋转轴；4、大姜姜种；5、插针；6、大姜鳞芽。
具体实施方式
以下结合附图对本发明的方法进一步描述，本实施例在以本发明技术方案为前提下进行实施，给出了详细的实施方式和具体的操作过程，但本发明的保护范围不限于下述的实施，例如姜种鳞芽对象。
参见图1-6，一种大姜精量定向种植中鳞芽识别及调整方法，包括HSV转化阶段、颜色分割阶段、图像二值化阶段、腐蚀操作阶段、膨胀操作阶段、图像识别阶段、偏差计算阶段以及鳞芽调整阶段，具体步骤如下：
第一步：HSV转化阶段，采集大姜姜种4的HSV颜色模型的图像或将大姜姜种4的其他彩色模型的图像转换为HSV颜色模型，并存储到内存中；如图2即为大姜姜种4的原始图像；
第二步：颜色分割阶段，对图像进行颜色分割操作，保留下图像中的大姜鳞芽6颜色区域；
第三步：图像二值化阶段，对保留下来的图像进行二值化；如图3所示；
第四步：腐蚀操作阶段，对二值化图像进行腐蚀操作；
第五步：膨胀操作阶段，对腐蚀后的二值化图像进行膨胀操作；如图4所示；
第六步：图像识别阶段，识别图像中大姜鳞芽6位置；
第七步：偏差计算阶段，计算大姜鳞芽6相对于0度基线的角度偏差；
第八步：鳞芽调整阶段，在种植时控制步进电机调整装置带着大姜姜种4旋转，实现大 姜鳞芽6方向的调整。
特别的，HSV颜色模型中颜色的参数分别是：色调H，饱和度S，明度V。
色调H，用角度度量，取值范围为0°～360°，从红色开始按逆时针方向计算，红色为0°，绿色为120°,蓝色为240°。它们的补色分别是：黄色为60°，青色为180°,品红为300°。
饱和度S，饱和度S表示颜色接近光谱色的程度。一种颜色，可以看成是某种光谱色与白色混合的结果。其中光谱色所占的比例愈大，颜色接近光谱色的程度就愈高，颜色的饱和度也就愈高。饱和度高，颜色则深而艳。光谱色的白光成分为0，饱和度达到最高。通常取值范围为0％～100％，值越大，颜色越饱和。
明度V，明度表示颜色明亮的程度，对于光源色，明度值与发光体的光亮度有关；对于物体色，此值和物体的透射比或反射比有关。通常取值范围为0％(黑)到100％(白)。
进一步的，所述颜色分割阶段是指通过设定色调H、饱和度S、明度V的特征参数，对大姜姜种4的HSV颜色模型的图像进行颜色分割操作，只保留颜色在[26，43，120]和[77，200，255]之间的像素点。上述两个矩阵范围内的色调H、饱和度S、明度V能将一般大姜鳞芽6的颜色完全包含入其中，所以通过该方法能非常快速的识别出姜芽位置。
进一步的，所述图像二值化阶段是指将颜色分割阶段留下的图像部分进行二值化操作，将图像转换为黑和白的图像。将整个图像呈现出明显的黑白效果，图像的二值化使图像中数据量大为减少，从而能凸显出目标的轮廓。二值化的阈值由OTSU方法(又称为最大类间方差法)计算得出，OTSU的中心思想是阈值T应使目标与背景两类的类间方差最大。对于一幅图像，设当前景与背景的分割阈值为t时，前景点占图像比例为w 0，均值为u 0，背景点占图像比例为w 1，均值为u 1，则整个图像的均值为u＝w 0×u 0+w 1×u 1。然后，建立目标函数g(t)＝w 0×(u 0-u) 2+w 1×(u 1-u) 2，则g(t)就是当分割阈值为t时的类间方差表达式。OTSU算法使得g(t)取得全局最大值，当g(t)为最大时所对应的t称为最佳阈值。
进一步的，所述腐蚀操作阶段是指对图像二值化阶段后得到的图像进行腐蚀操作，将二值化处理后图像中的噪点有效滤除而不破坏图像的原有信息，并提取出平滑的边缘。通常由于噪声的影响，图像二值化后的边界都很不平滑，物体区域具有一些噪声孔，而背景区域上散布着一些小的噪声物体。因此对二值化图像进行腐蚀操作，可以将上述步骤得到的图像中的噪点有效的滤除而不破坏图像的原有信息，该算法提取的边缘也比较平滑。
进一步的，所述膨胀操作阶段是指对腐蚀操作阶段后的图像进行膨胀处理，填充图像边界使图像边界变得更加平滑，让图像各区域更分明，同时不会明显改变图像区域原来的大小。
进一步的，所述图像识别阶段是指在膨胀操作阶段后得到的图像中寻找独立的颜色区域，计算各个区域的大小和位置，从而得到每片大姜鳞芽6的区域大小和位置(如图4所示，白色区域即为每片大姜鳞芽6的区域)，上述步骤已经将图像分为边界明显的大姜鳞芽6区域和背景区域，便于在图像中寻找独立的颜色区域。
所述偏差计算阶段是指用加权平均算法计算出一个新的大姜鳞芽6的位置A(如图5所示，黑色粗线左侧方框内的圆圈即为位置A)。然后以图像中心点为原点，以原点的0度基线为X轴，从图像中心画线到位置A的中心点，则该连线(如图5中黑色粗线所示)与0度基线的角度即为新的大姜鳞芽6的位置A与0度基线的角度偏差。
设f为每片大姜鳞芽6所占面积，x为每片大姜鳞芽6对应面积的中心点的位置矩阵，则新的大姜鳞芽6的位置A的位置矩阵
Figure PCTCN2020072126-appb-000001
则求得的x A的值即为新的大姜鳞芽6的位置A的中心点位置矩阵。
进一步的，所述鳞芽调整阶段是指在种植之前设定的基准位置，控制步进电机调整装置带动新的大姜鳞芽6的位置A旋转到与基准位置角度重合。
进一步的，所述步进电机调整装置包括步进电机1、设置于所述步进电机1下方的齿轮组2以及设置于所述齿轮组2下方的旋转轴3；所述齿轮组2包括相互啮合的第一齿轮和第二齿轮；所述第一齿轮与所述步进电机1的转轴同轴固定；所述第二齿轮与所述旋转轴3同轴固定；所述旋转轴3下方还放置有表面有大姜鳞芽6的大姜姜种4；有插针5上端与所述旋转轴3固定，所述插针5下端插入所述大姜姜种4；步进电机1驱动第一齿轮转动，通过第二齿轮能够带动旋转轴3转动，从而实现带动大姜姜种4上大姜鳞芽6方向的调整。
以上所述仅为本发明的实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。

Claims (9)

  1. 一种大姜精量定向种植中鳞芽识别及调整方法，其特征在于：包括HSV转化阶段、颜色分割阶段、图像二值化阶段、腐蚀操作阶段、膨胀操作阶段、图像识别阶段、偏差计算阶段以及鳞芽调整阶段，具体步骤如下：
    第一步：HSV转化阶段，采集大姜姜种(4)的HSV颜色模型的图像或将大姜姜种(4)的其他彩色模型的图像转换为HSV颜色模型，并存储到内存中；
    第二步：颜色分割阶段，对图像进行颜色分割操作，保留下图像中的大姜鳞芽(6)颜色区域；
    第三步：图像二值化阶段，对保留下来的图像进行二值化；
    第四步：腐蚀操作阶段，对二值化图像进行腐蚀操作；
    第五步：膨胀操作阶段，对腐蚀后的二值化图像进行膨胀操作；
    第六步：图像识别阶段，识别图像中大姜鳞芽(6)位置；
    第七步：偏差计算阶段，计算大姜鳞芽(6)相对于0度基线的角度偏差；
    第八步：鳞芽调整阶段，在种植时控制步进电机调整装置带着大姜姜种(4)旋转，实现大姜鳞芽(6)方向的调整。
  2. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述颜色分割阶段是指通过设定色调H、饱和度S、明度V的特征参数，对大姜姜种(4)的HSV颜色模型的图像进行颜色分割操作，只保留颜色在[26，43，120]和[77，200，255]之间的像素点。
  3. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述图像二值化阶段是指将颜色分割阶段留下的图像部分进行二值化操作，将图像转换为黑和白的图像。
  4. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述腐蚀操作阶段是指对图像二值化阶段后得到的图像进行腐蚀操作，将二值化处理后图像中的噪点有效滤除而不破坏图像的原有信息，并提取出平滑的边缘。
  5. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述膨胀操作阶段是指对腐蚀操作阶段后的图像进行膨胀处理，填充图像边界使图像边界变得平滑，让图像各区域更分明，同时不会明显改变图像区域原来的大小。
  6. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述图像识别阶段是指在膨胀操作阶段后得到的图像中寻找独立的颜色区域，计算各个区域的大小和位置，从而得到每片大姜鳞芽(6)的区域大小和位置。
  7. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述偏差计算阶段是指用加权平均算法计算出一个新的大姜鳞芽(6)的位置A，再计算位置A与0度基线的角度偏差。
  8. 根据权利要求1所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述鳞芽调整阶段是指在种植之前设定的基准位置，控制步进电机调整装置带动新的大姜鳞(6)芽的位置A旋转到与基准位置角度重合。
  9. 根据权利要求8所述大姜精量定向种植中鳞芽识别及调整方法，其特征在于：所述步进电机调整装置包括步进电机(1)、设置于所述步进电机(1)下方的齿轮组(2)以及设置于所述齿轮组(2)下方的旋转轴(3)；所述齿轮组(2)包括相互啮合的第一齿轮和第二齿轮；所述第一齿轮与所述步进电机(1)的转轴同轴固定；所述第二齿轮与所述旋转轴(3)同轴固定；所述旋转轴(3)下方还放置有表面有大姜鳞芽(6)的大姜姜种(4)；有插针(5)上端与所述旋转轴(3)固定，所述插针(5)下端插入所述大姜姜种(4)；步进电机(1)驱动第一齿轮转动，通过第二齿轮能够带动旋转轴(3)转动，从而实现带动大姜姜种(4)上大姜鳞芽(6)方向的调整。
PCT/CN2020/072126 2019-05-22 2020-01-15 一种大姜精量定向种植中鳞芽识别及调整方法 WO2020233156A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910427193.1A CN110178481B (zh) 2019-05-22 2019-05-22 一种大姜精量定向种植中鳞芽识别及调整方法
CN201910427193.1 2019-05-22

Publications (1)

Publication Number Publication Date
WO2020233156A1 true WO2020233156A1 (zh) 2020-11-26

Family

ID=67717223

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/072126 WO2020233156A1 (zh) 2019-05-22 2020-01-15 一种大姜精量定向种植中鳞芽识别及调整方法

Country Status (2)

Country Link
CN (1) CN110178481B (zh)
WO (1) WO2020233156A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114580511A (zh) * 2022-02-25 2022-06-03 南京中医药大学 一种基于图像亮度信息和投票机制的硫熏干姜鉴别方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110178481B (zh) * 2019-05-22 2020-08-21 青岛理工大学 一种大姜精量定向种植中鳞芽识别及调整方法
CN117876481A (zh) * 2023-12-22 2024-04-12 广东穗鑫高科智能科技有限公司 大蒜种子方向识别方法、装置、电子设备、存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007053933A (ja) * 2005-08-23 2007-03-08 National Agriculture & Food Research Organization 切り花の開花程度判断方法及び開花程度判断装置
CN101828449A (zh) * 2010-05-17 2010-09-15 高迟 大蒜蒜瓣鳞芽方向识别播种装置
CN102714960A (zh) * 2012-07-03 2012-10-10 河南省农业高新科技园有限公司 大蒜种粒鳞芽和根部识别装置
CN109104954A (zh) * 2018-08-22 2019-01-01 山东省农业机械科学研究院 大蒜定向投种机构及控制方法
CN109241965A (zh) * 2018-08-22 2019-01-18 山东省农业机械科学研究院 图像处理的大蒜鳞芽方向识别及作业监控装置和方法
CN110178481A (zh) * 2019-05-22 2019-08-30 青岛理工大学 一种大姜精量定向种植中鳞芽识别及调整方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI435234B (zh) * 2011-11-24 2014-04-21 Inst Information Industry Plant disease identification method, system and record media
CN104200193A (zh) * 2014-08-05 2014-12-10 北京农业信息技术研究中心 一种果树产量估测方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007053933A (ja) * 2005-08-23 2007-03-08 National Agriculture & Food Research Organization 切り花の開花程度判断方法及び開花程度判断装置
CN101828449A (zh) * 2010-05-17 2010-09-15 高迟 大蒜蒜瓣鳞芽方向识别播种装置
CN102714960A (zh) * 2012-07-03 2012-10-10 河南省农业高新科技园有限公司 大蒜种粒鳞芽和根部识别装置
CN109104954A (zh) * 2018-08-22 2019-01-01 山东省农业机械科学研究院 大蒜定向投种机构及控制方法
CN109241965A (zh) * 2018-08-22 2019-01-18 山东省农业机械科学研究院 图像处理的大蒜鳞芽方向识别及作业监控装置和方法
CN110178481A (zh) * 2019-05-22 2019-08-30 青岛理工大学 一种大姜精量定向种植中鳞芽识别及调整方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
曹世佳 (CAO, SHIJIA): "基于图像识别的种植机器人的研究与应用 (Research and application of planting robot based on image recognition)", 中国优秀硕士学位论文全文数据库 信息科技辑 I138-1870 (CHINA MASTER’S THESES FULL-TEXT DATABASE, INFORMATION SCIENCE & TECHNOLOGY, I138-1870), no. 2,, 15 February 2019 (2019-02-15), XP55756541, ISSN: 1674-0246, DOI: 20200305122447Y *
陆尚平 (LU, SHANGPING): "基于机器视觉的甘蔗茎节识别与蔗芽检测研究 (Research on sugarcane internodes and sugarcane buds identification based on machina vision)", 中国博士学位论文全文数据库 农业科技辑 D047-37 (CHINA DOCTORAL DISSERTATIONS FULL-TEXT DATABASE, AGRICULTURE SCIENCE AND TECHNOLOGY, D047-37), no. 5, 15 May 2015 (2015-05-15), XP55756545, ISSN: 1674-022X, DOI: 20200305122435Y *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114580511A (zh) * 2022-02-25 2022-06-03 南京中医药大学 一种基于图像亮度信息和投票机制的硫熏干姜鉴别方法
CN114580511B (zh) * 2022-02-25 2024-05-24 南京中医药大学 一种基于图像亮度信息和投票机制的硫熏干姜鉴别方法

Also Published As

Publication number Publication date
CN110178481B (zh) 2020-08-21
CN110178481A (zh) 2019-08-30

Similar Documents

Publication Publication Date Title
WO2020233156A1 (zh) 一种大姜精量定向种植中鳞芽识别及调整方法
CN108009542B (zh) 油菜大田环境下杂草图像分割方法
CN109711325B (zh) 一种芒果采摘点识别方法
CN107895376A (zh) 基于改进Canny算子和轮廓面积阈值的太阳能电池板识别方法
CN109255757B (zh) 一种机器视觉自然放置葡萄串果梗区域分割方法
CN111666815B (zh) 一种基于Sentinel-2遥感影像的大蒜种植信息自动化提取方法
CN107038446A (zh) 一种主动照明下基于重叠边缘检测的夜间双果重叠番茄识别方法
CN105719282B (zh) 一种果园红色苹果图像果实枝叶区域获取方法
CN103336966A (zh) 一种应用于农业智能机械的杂草图像辨识方法
CN111784764A (zh) 一种茶叶嫩芽识别与定位算法
CN102542560B (zh) 一种水稻移栽后密度自动检测的方法
CN114708208B (zh) 一种基于机器视觉的名优茶嫩芽识别与采摘点定位方法
CN114049365A (zh) 一种双孢蘑菇视觉分割识别方法
CN107085851A (zh) 一种提取水域岸线的方法及装置
CN108037123A (zh) 一种杂交稻钵体盘播种性能参数精密检测方法
CN108805875B (zh) 双孢菇位置识别与推倒方向的选定方法
CN112329506A (zh) 枸杞采摘机器人的果实识别方法及系统、定位方法及系统
CN102339378A (zh) 一种自动提取棉苗的方法及其装置
CN104636716B (zh) 绿色果实识别方法
CN115294482B (zh) 一种基于无人机遥感图像的食用菌产量估算方法
CN110223339A (zh) 一种基于机器视觉热保护器校准点中心定位方法
CN114240758B (zh) 一种以四边形地块为参照物的山区茶园低空图像拼接方法
CN107169940B (zh) 基于电子识别的单株梨树产量获取方法
CN107657617A (zh) 一种红花花丝识别方法
CN114581660A (zh) 植物叶片分割识别方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20808894

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20808894

Country of ref document: EP

Kind code of ref document: A1