Qualityassessmentofcorngrainsampleusingcolorimageanalysis(1)资源-CSDN文库

需积分: 9 45 浏览量 2015-03-25 15:25:19 上传评论收藏 614KB PDF 举报

### 使用颜色图像分析评估玉米粒质量的关键知识点 #### 标题与描述中的核心知识点解析 **标题与描述**:“使用颜色图像分析评估玉米粒样本的质量（1）” 此标题及描述明确指出了研究的主要目标——利用颜色图像分析技术对玉米粒的质量进行客观评估。这一方法旨在通过计算机视觉手段，代替或辅助人工专家的传统视觉评估方式，从而提高评估过程的准确性与一致性。 #### 关键词解析 - **谷物质量评估**：指的是根据谷物的各项特性如外观、形状、颜色、气味、风味、含水量、感染情况以及杂质含量等来评定其品质的过程。 - **计算机视觉**：一种让计算机能够理解并解释图像的技术，本研究中用于提取玉米粒的颜色和形状特征。 - **特征提取**：从原始图像数据中提取出有助于识别和分类的重要信息的过程。 - **分类**：将提取到的特征映射到预定义类别中的过程，在本文中是指将玉米粒归类到不同的质量等级。 #### 内容部分的核心知识点 ##### 研究背景 - **谷物质量评价的重要性**：谷物作为广泛食品的重要组成部分，其质量直接影响食品安全和营养价值。传统的质量评估依赖于专家的经验，存在主观性及不一致性问题。 - **颜色和形状在质量评估中的作用**：颜色和形状是评判谷物质量的关键指标之一。颜色可以反映谷物的新鲜度和可能存在的病害情况；形状则有助于识别是否有损伤或异形粒的存在。 ##### 方法论 - **基于颜色图像分析的评估方法**：该研究采用了一种基于颜色图像分析的方法，通过对玉米粒的图像进行复杂分析来实现质量评估。 - **特征提取与描述**：研究中提出了用于特征提取的方法，包括颜色和形状特征，并为被评估对象提供了描述性的表征。 - **分类器设计**：开发了三个基于径向基函数（Radial Basis Function, RBF）的分类器，用于识别玉米粒的颜色和形状类别。 - **结果融合**：为了综合颜色和形状两个方面的评估结果，研究探讨了两种不同的融合策略。 - **性能评估**：通过计算训练误差和测试误差来评估所开发流程的有效性和可靠性。 ##### 结论通过上述方法的应用，本研究提供了一种更加客观且可靠的玉米粒质量评估手段。这种方法不仅可以减少人为因素带来的不确定性，还能够提高评估效率，对于保障食品安全和提升农产品加工过程中的质量控制具有重要意义。该论文介绍了一套完整的基于颜色图像分析的玉米粒质量评估体系，从理论基础到具体实践均进行了详细的阐述，为后续的相关研究提供了有价值的参考。

资源推荐

资源详情

资源评论

ORIGINAL PAPER

Quality assessment of corn grain sample using color image

analysis

Miroljub Ivanov Mladenov

•

Stanislav Miroslavov Penchev

•

Martin Plamenov Dejanov

•

Metin Sebahatin Mustafa

Received: 18 April 2011 / Accepted: 1 October 2011 / Published online: 30 October 2011

 Springer Science+Business Media, LLC 2011

Abstract Grain quality is assessed based on different

grain features like appearance, shape, color, smell, ﬂavour,

moisture content, infections, presence of impurities, etc.

The main indexes for the quality of grain samples are

related to the color characteristics and the shape of the

grain sample elements. Most of these characteristics are

assessed visually by an expert. In this paper, an approach

for an objective estimation of some basic grain quality

characteristics is presented. It is based on a complex

analysis of color images of the investigated objects. Due to

the conceptual difference in presenting the objects’ color

and shape characteristics, their assessment was performed

separately. After that the results form these two assess-

ments were combined and the ﬁnal decision about the

object’s classiﬁcation to one of the quality groups deﬁned

by the standard regulations was made. Methods and tools

for feature extraction and for object description, as well as

for classiﬁcation of the objects into predetermined groups

were proposed. Three classiﬁers, based on radial basis

elements, which were used for grain color and shape class

recognition, were analyzed. Two different approaches for

fusing the results from object color and object shape

analyses were investigated. The training and testing errors

of the developed procedures were evaluated.

Keywords Grain quality assessment  Computer vision 

Feature extraction  Classiﬁcation

Introduction

The grain is one of the agricultural products, which is a main

part of a wide range of foods. The assessment of grain quality

is based on the evaluation of different characteristics like

appearance, shape, color, smell, ﬂavour, moisture content,

infections, presence of impurities etc. An expert, using visual

estimation only, assesses part of these characteristics (for

example appearance, shape, color, presence of impurities,

etc.). The quality of such assessment depends on the expert’s

knowledge, experience and capability of result interpreta-

tion. This deﬁnes the nature of the assessment—subjective,

slow and insufﬁciently precise. That is why the development

of new advanced, objective and automated technologies for

grain quality and safety assessment is the main goal of a wide

range of investigations and research programs.

Such is one of the main objectives of the project

‘‘Development of Intelligent Technologies for Assessment

of Quality and Safety of Food Agricultural Products’’

(INTECHN) [1], funded by the Bulgarian National Science

Fund. Within the frames of the project, intelligent tech-

nologies and tools for assessment of the quality and safety

of widespread food products, like cereals, fruits, vegeta-

bles, milk, dairy products, meat and meat products and

eggs were developed.

The INTECHN project idea is to investigate the possi-

bility for complex assessment of some basic grain quality

characteristics by analysis of both color images and spec-

tral characteristics of the grain sample elements. Using

color image analysis, we can obtain data about visible

features, related to the color characteristics, shape and the

dimensions of the objects.

In this paper the approaches, methods and tools for

quality assessment of corn grain samples, as well as part of

the results obtained are presented.

M. I. Mladenov  S. M. Penchev (&)  M. P. Dejanov 

M. S. Mustafa

Department of Automatics, Information and Control

Engineering, University of Ruse, 8 Studentska str. 7017,

Ruse, Bulgaria

e-mail: msp@uni-ruse.bg

123

Sens. & Instrumen. Food Qual. (2011) 5:111–127

DOI 10.1007/s11694-011-9118-4

According to the Bulgarian Government Standards [2,

3], the main characteristics of grain quality are appearance,

shape, color, smell, ﬂavour, moisture content, presence of

impurities—grain and non-grain ones. Whole grains with

appearance, shape and color inherent for the variety and

hybrid, as well as broken grains bigger than the half of the

whole grain, are considered to be standard. There are

several groups, which are considered as grain impurities:

broken grains smaller than the half of the whole grain;

heat-damaged grains, burned grains with damaged kernels;

small, shriveled and green grains; sprouted grains, grains

with moldy germs, dark brown to black; Fusarium infected

grains. The group of the non-grain impurities consists of:

damaged grains, corn-cob particles, leaf and stem fractions,

pebbles, soil, sand, smutty grains, as well as harmful ele-

ments (bunt).

Most of the grain quality characteristics mentioned

above (except moisture content, smell and ﬂavour) are

related to the color characteristics, shape and dimensions of

the grain sample elements. The main trend in the last few

years has been to use Computer Vision Systems (CVS) in

order to evaluate such characteristics. A review of the

progress of computer vision in the agricultural and food

industry was given in [4]. To obtain a complex assessment

of the grain quality using data about color characteristics,

shape and dimensions of the grain sample elements, is a

complicated and multilevel task. This is because the color

characteristics shape and dimensions of the elements in a

sample vary within a wide range.

Many results have been published, in which color char-

acteristics analysis was used to assess some particular quality

features. A digital image analysis algorithm, developed to

facilitate classiﬁcation of individual kernels of Canada

Western Red Spring (CWRS) wheat, Canada Western

Amber Durum (CWAD) wheat, barley, oats, and rye using

textural features of individual grains was reported in [5]. The

textural features of individual kernels were extracted from

different colors and color band combinations of images to

determine the color or the color band combination that gave

the highest classiﬁcation accuracies in assessment of

authenticity of cereal grains. Color characteristics analysis

was used also to assess class [5–7], infections [8, 9], ger-

mination [10], weed identiﬁcation [11], etc.

Morphological features, related to the grain shape and

geometrical parameters were used for assessment of the

grain variety. A set of eight morphological features

namely, area, perimeter, length of major axis, length of

minor axis, elongation, roundness, Feret diameter and

compactness to recognize ﬁve different kinds of cereal

grains was presented in [12]. A broader investigation, with

a total of 230 features (51 morphological, 123 color, and 56

textural) used for classiﬁcation of barley, Canada Western

Amber Durum wheat, Canada Western Red Spring wheat,

oats, and rye was presented in [13]. Assessment of the

grain sample purity was performed by proﬁle analysis of

corn kernels using one-dimensional digital signals based on

its binary images [14], by modeling the shape using a set of

morphological features [15] and by shape curvature anal-

ysis [16]. Computer vision methods were also used for

determination of different injuries: corn (Zea mays) kernel

mechanical damage and mold damage [17], whole and

broken kernel identiﬁcation [18], etc.

It is very important to use proper classiﬁcation proce-

dures in order to assess precisely the main features of grain

sample elements. Different approaches were used for this

purpose. Some effective results were obtained using clas-

siﬁcation based on multi-layer neural networks (NN) [19].

An evaluation of the classiﬁcation accuracy of nine dif-

ferent neural network architectures to classify ﬁve different

kinds of cereal grains was performed in [12]. A comparison

between a four-layer backpropagation neural network and a

non-parametric statistical classiﬁer when solving a similar

task was presented in [15]. The results show that the neural

network classiﬁer outperformed the non-parametric clas-

siﬁer in almost all the instances of classiﬁcation.

Neural Networks were applied for quality analysis of

grain sample elements in combination with color charac-

teristics [13, 20], shape and dimensions of the objects [12,

16, 20]. They were used for evaluation of the class [12, 15],

purity [14, 16], different kernel mechanical damages [20],

germination [10], infections [7, 20], weed seeds detection,

etc.

The references cited above present some results con-

cerning the assessment of speciﬁc grain quality features.

The main goal of this paper is to present the INTECHN

approach for a complex assessment of the corn grain

quality. This approach is based on the analysis of the object

color characteristics, shape and dimensions using CVS, as

well as the fusion of the results of the separate analyses.

The ﬁnal goal of the complex assessment is the categori-

zation of the grain sample elements in three quality groups.

The INTECHN platform approaches, methods and tools for

recognition of the color characteristics and shape of grain

sample elements, as well as the approaches for fusing the

results from color and shape analysis, are presented. Some

of the results obtained at this stage of project implemen-

tation are given too.

Materials and methods

INTECHN platform hardware

The INTECHN platform CVS (Fig. 1) includes two color

CCD cameras (DFK31AU03, THE IMAGINGSOURCE,

Germany) (1) with lenses (PENTAX B2514D, Hoya

112 M. I. Mladenov et al.

123

Corporation, Japan) (3), which give a possibility to form

color images of the investigated object (2) in two planes

(horizontal and vertical). The illumination system includes

two luminescent ring-shaped units (4), which have different

diameters. It is used for direct object illumination. A portable

computer model Dell Vostro 1720 was used for implemen-

tation of the INTECHN platform software procedures.

Image acquisition

The images of individual kernels were taken in uniform

illumination conditions (see Fig. 1), the cameras were

preliminary calibrated using the software provided by the

manufacturer. The ﬁles were stored in RGB24 color space

with spatial resolution of 1024 9 768 pixels. The software

procedures for image acquisition, processing and analysis

were developed using MATLAB (Version 7, The Math-

works, Inc., Natick, MA).

When the object was inserted in the ﬁeld of view of the

cameras, its color images appeared in the two windows

(‘‘Upper camera’’ and ‘‘Side camera’’) of the Graphical User

Interface (GUI), presented in Fig. 2. In most cases, the image

from the upper camera was analyzed, but sometimes when

some parts of the object (broken parts, the germ, etc.) are not

visible, the operator can choose the image from the side

camera. The selected image was loaded by clicking on the

button ‘‘Browse’’ and selecting the respective ﬁle.

Grain sample classiﬁcation groups

Within the frames of the INTECHN project, CVS was used

to assess such characteristics of grain sample elements,

which are usually evaluated by an expert based on visual

estimation only. Groups (classes) and subgroups (subclas-

ses), in which the corn grain sample elements have to be

distributed, are presented in Tables 1, 2, 3.

It is evident from the tables, that the main characteris-

tics, which determine the grouping of the grain sample

elements in classes and subclasses, are related to the

objects’ color and shape. Because of the fact that color and

shape features are described in a different manner, it is

advisable the assessment of these characteristics to be

made separately. After that, the results from the two

assessments have to be combined to obtain the result about

the object’s classiﬁcation in one of the normative classes.

In terms of the classiﬁcation procedure, color and shape

groups were divided into several subgroups. Color char-

acteristics were separated in eight basic classes (color

classes), which correspond to the color zones typical for

different sample elements, and one additional class, which

corresponds to the non-grain impurities (it is impossible to

deﬁne a compact class for non-grain impurities) (Table 2).

In terms of the shape, the objects were divided into three

basic classes (shape classes), which correspond to the

whole grains, broken grains bigger than the half of the

whole grain and broken grains smaller than the half of

the whole grain, and one additional class, which corre-

sponds to the non grain impurities (Table 3).

Corn grain samples of the Kneja—433 variety were used

in the investigation. The Maize Research Institute-Kneja,

Bulgaria, produces this variety. The samples were gathered

in one corn-growing season of one crop year and from one

growing location. Principally, the INTECHN platform

presumes the grain kind, variety and class to be speciﬁed

when we develop the database of the grain color charac-

teristics and grain shapes. There is a possibility for the

growing location to be speciﬁed too. Since the grains of

different varieties have different color characteristics and

shapes, a new database has to be developed for every new

variety. An analysis how the growing region inﬂuences the

corn grain color and shape, was not made at this stage of

the project implementation.

The Fusarium infection was selected because such

infection is often present in corn crops. If we have to detect

another kind of infection, which has visible features on the

grain surface, we have to develop a new color zone model.

The model has to correspond to the pixels from the infected

area, which have color characteristics speciﬁc for the

infection.

Determination of the color class of the grain sample

elements

The analysis of the color characteristics of the corn grain

sample elements showed, that the following typical color

zones could be found in different sample element images:

1color zone (cz)—grain back side zone, 2cz—grain germ

zone, 3cz—heat-damaged grain zone, 4cz—green grain

zone, 5cz—mold grain zone, 6cz—smutty grain zone,

Fig. 1 INTECHN platform computer vision system

Quality assessment of corn grain sample 113

123

剩余16页未读，继续阅读

评论收藏

内容反馈

sinat_26865803

粉丝: 0
资源: 2

Quality assessment of corn grain sample using color imageanalysi...

最新资源

Quality assessment of corn grain sample using color imageanalysi...

using color

ImageAnalysis1

Methodology for the subjective assessment of the quality of television pictures.dpf

A Model for Data Quality Assessment .pdf

data quality assessment

Screen Content Image Quality Assessment Using Multi-Scale Difference of Gaussian

blind image quality assessment for measuring image blur

Quality assessment for visible and infrared color fusion images of typical scenes

Visual Quality Assessment for Natural and Medical Image(2018).pdf

SAE EIA 554-1 Assessment of Average Outgoing Quality Levels in

2016 infovis 论文集

image_quality_assessment.zip_quality assessment_人眼 质量_图像 评估_图像质量

iqa 源代码 （Image Quality Assessment 图像质量评价库）

Modern Image Quality Assessment_Zhou Wang_Alan C. Bovik

Blind Image Quality Assessment

Learning to rank for blind image quality

image quality assessment经典论文

Using binocular feature combination for blind quality assessment of stereoscopic images

Quality of life assessment of children with thalassemia.pdf

Modular Blind Video Quality Assessment

NIMA Neural Image Assessment (TIP 2018).pdf

C3DVQA: FULL-REFERENCE VIDEO QUALITY ASSESSMENT WITH 3D CONVOLUT

9.Quality-Assessment-of-Deblocked-Images.rar_After This_quality

2.Liu_RankIQA_Learning_From_ICCV_2017_paper.pdf

最新资源

image_quality_assessment.zip_quality assessment_人眼质量_图像评估_图像质量

iqa 源代码（Image Quality Assessment 图像质量评价库）