212

Journal homepage: www.fia.usv.ro/fiajournal 
Journal of Faculty of Food Engineering,  

Ştefan cel Mare University of Suceava, Romania  
Volume XIV, Issue 2 - 2015, pag. 212 - 217 

 

 

DEVELOPMENT OF OPERATIONAL QUALITY CONTROL METHOD FOR 

MEAT PRODUCTS 

 
Olha LYUBCHYK1, Mykola MYKYJCHUK1, *Mariуa VOROBETS2  

1National University “Lviv Polytechnic”, Department of Metrology, Standardization and Certification, 12, 
S.Bandery str., Lviv, Ukraine, 79013  

2Yury Fedkovych National University of Chernivtsi, Ukraine, m.vorobets@chnu.edu.ua 
* Corresponding author 

Received March 14th 2015, accepted June 17th 2015 

 
Abstract: Assortment of finished products of meat in the consumer market in Ukraine is quite wide, 
but a large number of products are very varied in quality. Today quality control of meat and meat 
products is very important. Doing assessing of a quality of meat, the importance is focused on meat 
organoleptic characteristics which shall identify the most characteristic features of a particular type 
of meat for the consumer and classify them. The following distinguishing features of meat freshness 
are: the number of fibers, medium size, color, edge definition fibers. A method for operational quality 
control for meat products by customers in the points of product sale based on the optical methods 
using modern digital optoelectronic has been developed. In the article are described problems that 
shall be solved to implement optical method of operational identification of a meat type. The 
operational analysis method allows getting fast results and assesses the structure of most types of 
meat products, which gives reason to recommend it for quality control and detect adulteration of meat 
products. 
 
Keywords: pattern recognition, adulteration, microstructural analysis, meat products, operational 
quality control 
 
 
1. Introduction 
 
Today a consumer has an opportunity to 
choose a product from a wide range of 
goods with the same (or similar) consumer 
properties, but here he encounters another 
problem – how to evaluate a product 
quality with the optimal properties out of 
this range. Under these conditions of 
commodity sufficiency the consumer 
should be provided with as much 
information as possible to enable him to 
identify the product properly and buy the 
one quality product that will satisfy his 
needs to the fullest extent.  
An important factor in assurance of the 
required quality of the food products is the 
availability of the objective information 

about the quality parameters obtained by 
measurements using appropriate methods 
and techniques. The correct choice of the 
measurement methods and techniques is 
essential to ensure the information 
adequacy on the actual level of the food 
product quality. In determining the quality 
of meat the most typical features of a 
particular type of meat shall be identified  
for the consumer, classified, for example, 
using the pattern identification theory and 
up-to-date programming units.  
The operational impact on the production 
process of meat and meat products shall be 
an objective of the modern process 
organization in order to minimize the 
consumer and environmental risks. 
 

http://www.fia.usv.ro/fiajournal
mailto:m.vorobets@chnu.edu.ua


Food and Environment Safety - Journal of Faculty of Food Engineering, Ştefan cel Mare University - Suceava 
Volume XIV, Issue 2 – 2015 

 

Olha LYUBCHYK, Mykola MYKYJCHUK, Mariуa VOROBETS, Development of operational quality control method 
for meat products, Food and Environment Safety, Volume XIV, Issue 2 – 2015, pag. 212 – 217 

213 
 

2. Materials and Methods 
 
The range and sales of meat products have 
significantly increased over the recent 
years. In the meat market, being in the 
steady demand of consumers, its various 
types are available and it is sometimes 
hard to the customer to choose a quality 
product out of this variety.  
Increase in the production rates and 
volumes of output of the meat industry 
requires improvement of existing and 
development of new production processes, 
which ensure efficient use of raw material 
resources, increase in the output and 
improvement in the product quality.  
The following parameters play the main 
role in assessing the quality of meat:  
• contents of ingredients used by the body 
for biological synthesis and covering the 
metabolic cost;  
• organoleptic characteristics (appearance, 
odor, color, texture);  
• absence of toxic substances and 
pathogens.  
According to Law of Ukraine “On the 
Safety and Quality of Food Products” all 
food products getting to the consumer 
market shall pass the safety check during 
the compulsory certification. It is 
practically impossible to detect 
adulteration using conventional methods, 
because one has to check not only the type 
and grade of products, but also to carry out 
their identification by ingredients making 
it possible to identify and verify the 
adequacy of the information specified on 
the label, compliance with the formulation, 
type and name of the product [1].  
Enzyme immunoassay methods, 
chromatographic methods and PCR 
diagnostics are used most of all. They are 
highly sensitive, accurate and fast. 
However, they are difficult to adapt to such 
complicated multicomponent systems as 
meat products and with all the advantages 
of these methods they are not able to detect 
substitution of meat with other animal 

ingredients – lacteal gland, liver, vascular 
glands and other by-products. Application 
of chemical, physicochemical and 
biochemical methods provides information 
about the energy value of meat products 
[1].  
The existing control methods shall be 
improved towards their cheapening and 
approaching the point of sale to increase 
the reliability and efficiency of 
identification of the meat quality. 
 
3. Results and Discussion 
 
The range of end products of meat is quite 
wide in the consumer market in Ukraine, 
but a large number of products vary in 
quality. As studies show [2], in many cases 
while producing food items from meat, 
producers substitute meat with the 
phytogenic products, including soy protein, 
introduce low value additives, which were 
not provided by formula, and often use 
tainted meat or reused materials. Quality 
control of meat and meat products is the 
most topical issue today. The control 
methods in Ukraine are obsolete and do 
not enable to determine all modern 
adulterations. Forthis purpose many 
European countries such as Germany, 
Austria, and Russia apply microstructure 
analysis of meat products based on the 
principles of histopathological and 
histochemical examinations. 
Method of microstructural analysis is a 
methodological approach to the study of 
short-term cognitive and executive actions. 
It is assumed that in each time interval 
after receiving instructions for the 
execution of action – specific input data 
conversion happens due to inclusion of a 
particular functional unit [3]. 
Meat products at various stages of 
processing treatment, as well as finished 
products retain their morphological 
features. Therefore, using microstructure 
analysis of raw materials, semi-finished or 
finished products, we can determine the 



Food and Environment Safety - Journal of Faculty of Food Engineering, Ştefan cel Mare University - Suceava 
Volume XIV, Issue 2 – 2015 

 

Olha LYUBCHYK, Mykola MYKYJCHUK, Mariуa VOROBETS, Development of operational quality control method 
for meat products, Food and Environment Safety, Volume XIV, Issue 2 – 2015, pag. 212 – 217 

214 
 

presence of certain types of tissues, organs, 
spices, and low-grade additives [2, 4–6]. 
The microstructure analysis makes it 
possible not only to detect adulteration but 
also to control the compliance of the 

ingredients of the meat products with the 
approved formula [7]. 
The conclusion on the meat freshness is 
drawn based on the test results [8], 
according to the distinguishing features 
shown in Table 1.  

 Table 1 
Organoleptic indicators of meat of different grade of freshness 

Indicator 
Characteristic feature of meat 

Fresh Doubtful freshness Tainted  
Color and 
appearance of 
product 

Drying crust is pale pink or 
pale red 

Wet in places, slightly 
sticky, darkened 

Very dried, covered with 
grayish-brown slime or mold 

State of fat 
 
 
 
 
 
 

Beef has a white, yellowish 
or yellow color; firm 
texture, crumbles when 
squeezing 
Pork has a white or pale 
pink color, soft, elastic 
Chicken has a pale yellow 
or yellow color 

Has a grayish-matte shade, 
slightly sticks to fingers 
 
 
 
 
Chicken has a pale yellow 
or yellow color 

Has a grayish-matte shade, 
smears when squeezing.  
Pork fat can be covered with 
a bit of mold. The smell is 
rancid (expresses bitterness). 
Chicken fat has a pale white 
color with a gray shade 

Muscles in the cut Slightly wet, do not leave 
wet spots on filter paper; 
color is common to  this 
type of meat 

Wet, leave wet spots on 
filter paper, are slightly 
sticky; have a dark red 
color 

Wet, leave wet spots on filter 
paper, are sticky; have a red-
brown color 

Texture  The meat is dense and 
elastic in the cut; flexible, 
when pressing with a finger 
a dent appears, which is 
quickly recovered 

The meat is less dense in 
the cut, when pressing with 
a finger a dent appears, 
which is slowly recovered 

The meat is flabby in the cut, 
when pressing with a finger 
a dent appears, which is not 
recovered  

Odor Peculiar, specific to each 
type of fresh meat 

Slightly acid with a hint of 
mustiness 

Acid or musty or slightly 
odorous 

State of tendon Tendons are elastic, dense, 
articular surfaces are 
smooth, shiny 

Tendons are less dense, 
have a dull white color. 
Articular surfaces are 
slightly covered with slime 

Tendons are softened, have a 
grayish color. Articular 
surfaces are covered with 
slime 

 
The choice of the characteristic features 
was made based on the study results 
specified in [1, 2] and on their control 
availability using simple optical devices 
(digital cameras). 
For the application of the proposed 
approach for the operational control of the 
meat quality, it is necessary to identify the 
most characteristic features of a particular 
type of meat for the consumer and classify 
them, for example, using pattern 
recognition theory and up-to-date 
programming units. A pattern recognition 
system involves the following three 

aspects [9]: data acquisition and 
preprocessing; data representation; 
decision making. 
Pattern recognition presents one of the 
most significant challenges for scientists 
and engineers, and many different 
approaches have been proposed [10]. 
Within the artificial intelligence theory 
pattern recognition is included in the 
broader science discipline – computer-
assisted instruction theory, which purpose 
is to develop algorithm design methods 
able to learn. Machine pattern recognition 
mainly has two basic methods: statistics 



Food and Environment Safety - Journal of Faculty of Food Engineering, Ştefan cel Mare University - Suceava 
Volume XIV, Issue 2 – 2015 

 

Olha LYUBCHYK, Mykola MYKYJCHUK, Mariуa VOROBETS, Development of operational quality control method 
for meat products, Food and Environment Safety, Volume XIV, Issue 2 – 2015, pag. 212 – 217 

215 
 

pattern recognition and structure pattern 
recognition [11]. 
Pattern recognition is widely applicable 
and is used in the creation of all computer 
systems having intellectual functions, i.e. 
functions related to decision making 
instead of a human: computer-aided 
diagnosis, criminalistics expert 
examination, information search and data 
mining, etc. 
Each image is a set of numbers that 
describe the properties and are called 
attributes. Ordered set of attributes of an 
object is called feature vectors. Feature 
vector is a point in the space of features. 
The classifier or the decision rule is a 
function that assigns a class to the feature 
vector pattern, to which it belongs. 
The problem of pattern recognition can be 
divided into a number of subtasks. 
1. Generation of signs - measurement or 
calculation of numerical features that 
characterize the object. 
2. Selecting features - identifying the most 
informative features for classification (both 
early signs, and their functions can be 
included in this set). 
3. Building the classifier formation of the 
decision rule, under which the 
classification is made. 
4. Evaluation of the classification quality - 
calculation of the rate of classification 
accuracy (accuracy, sensitivity, specificity, 
first and second kind errors). 
Let us introduce the notation and formulate 
a mathematical classification problem. 
We use the following classification 
problem model [12]:  
Ω – a set of recognition objects (pattern 
space).  
ω: ωΩ – recognition object (pattern).  
g (ω): Ω →M, M={1, 2, ..., m} – indicator 
function that breaks up the pattern space 
into Ω to m of disjoined classes Ω1, Ω2, ..., 
Ωm. Indicator function is unknown to the 
observer.  
X – space of observations perceived by the 
observer (feature space).  

x (ω): Ω → X – function that assigns to 
each object ω point x (ω) in the feature 
space. Vector x (ω) is the pattern of the 
object perceived by the observer. In the 
feature space disjoint set of points is 
determined Ki∩X, i=1, 2 ..., m, 
corresponding samples of one class.  
gˆ(x): X → M – decision rule – estimation 
for g(ω) based on x(ω), that is gˆ(x) = 
gˆ(x(ω)).  
Let (xj)=x(ω j), j=1, 2 ..., N – the 
information about function g(ω) and x(ω) 
be available the observer but most of these 
features are unknown to the observer. Then 
(gj, xj), j=1, 2 ..., N – there are many 
precedents.  
The problem is to construct such decision 
rule gˆ(x) to make the recognition with the 
minimum number of errors.  
The usual case - to consider the feature 
space as Euclidean of X = R1. The quality 
of decision rule is measured with the 
frequency of proper solutions. Usually it is 
estimated by providing a certain 
probability measure to a set of objects Ω. 
Then the problem is written as min 
P{gˆ(x(ω)) ≠ g(ω)}. 
The following problems shall be solved to 
implement optical method of operational 
identification of a type of meat:  
1) determine the characteristic features of 
a particular type of meat based on a 
number of pictures of different types of 
meat;  
2) create a mathematical model of types of 
meat that will consider its characteristic 
features (virtual standards of types of 
meat);  
3) develop algorithms for operational 
identification, which based on the pattern 
recognition theory will quickly identify a 
type of meat and its level of quality by a 
consumer himself. 
Operational quality control method for 
meat products was developed based on the 
analysis made, shown in Figure 1. 
At the first stage we determine 
identification characteristics of different 



Food and Environment Safety - Journal of Faculty of Food Engineering, Ştefan cel Mare University - Suceava 
Volume XIV, Issue 2 – 2015 

 

Olha LYUBCHYK, Mykola MYKYJCHUK, Mariуa VOROBETS, Development of operational quality control method 
for meat products, Food and Environment Safety, Volume XIV, Issue 2 – 2015, pag. 212 – 217 

216 
 

types of meat: number of fibers, average 
fiber size, color, and edge definition of 
fibers. The color of meat is one of the main 
quality indicators, estimated by a 
consumer, by which a marketable 
appearance of products is judged, as well 
as some chemical transformations, which 
may take place in the meat. Muscular 
tissue is the main edible part, which 

consists of separate long thin fibers 
covered with a thin translucent membrane. 
The basis of connective tissue is collagen 
and elastin fibers. There are following 
types of connective tissue depending on 
their ratio and location: loose, dense, 
elastin and reticular. 
 

 

 
 

Fig.1. Algorithm of the operational quality control for meat products
  
In order to identify the type of the meat 
samples, one should determine their 
statistical characteristics of the meat 
features, such as mathematical expectation 
and dispersion.  
According to the study it is necessary to 
create a classifier of types of meat, which 
shall include a database with statistical 
characteristics of meat features 
classification. This classification is shown 
in Table 2. 

Its statistical characteristics are determined 
based on the results of estimate of the meat 
classification features and comparison is 
made by the features being in the classifier. 
The meat under examination is identified 
when comparing the characteristics and 
probability estimate is made by the 
decision rule. 
  

 
 
 

Choice of 
classification 

features of a sample 
of meat 

Determination of 
statistical 

characteristics of a 
sample of meat 

Construction of the 
classifier of types of 

meat 

Measurement of 
classification 

features of a sample 
of meat 

Determination of 
statistical 

characteristics of a 
sample of meat 

 

Identification of the 
type of meat (by the 

decision rule) 

 
Comparison of the 

meat characteristics by 
the classification 

features 

Estimate of the 
identification 
probability 



Food and Environment Safety - Journal of Faculty of Food Engineering, Ştefan cel Mare University - Suceava 
Volume XIV, Issue 2 – 2015 

 

Olha LYUBCHYK, Mykola MYKYJCHUK, Mariуa VOROBETS, Development of operational quality control method 
for meat products, Food and Environment Safety, Volume XIV, Issue 2 – 2015, pag. 212 – 217 

217 
 

Table 2 
Meat features classification  

Features Pork Beef Chicken 
Number of fibers nc nт nк 

 Average fiber size dс dт dк 
Color pale-pink to red pale-pink to dark red pale pink to pink 

 Edge definition of fibers ŋс ŋт ŋк 

 
4. Conclusions  
 
Adulteration of meat not only has an effect 
on the lower quality of the finished 
products, but it may be a hazardous factor 
for consumers’ health. 
Operational analysis method enables not to 
conduct additional analyses to obtain 
characteristics of the sample under 
examination. This enables to use this 
method in different operating conditions 
and facilitates the analysis. 
Use of operational analysis enables to 
identify adulteration, establish a grade of 
freshness and meat composition, its 
substitution with various ingredients. 
Today there is an urgent need for means of 
quality control of meat products in the 
places of distribution. These facilities must 
meet two basic requirements: to provide a 
sufficient level reliability of information on 
the status of quality meat and have a low 
price that will contribute to mass use. 
 
5. References 
 
[1]. KOTSYUMBAS I.Y., KOTSYUMBAS G.I., 
SCHEBENTOVSKA O.M., Examination of semi-
finished meat and meat-vegetable products using 
microstructure method, Recommended practice. 
Lviv: Afisha, 80, (2011). 
[2]. KOTSYUMBAS G.I., BISYUK I.U., 
KOTSYUMBAS I.Y., SCHEBENTOVSKA O.M., 

RUDYK G.V., MYSIV O.V., KOZAK M.V., 
Microstructure study of raw materials in ground 
meat, Recommended practice, Lviv: Afisha, 48, 
(2006). 
[3]. ZINCHENKO V.P., Image and activities, 
Institute of Practical Psychology, Voronezh: NPO 
MODEK, 608, (1997). 
[4]. ADUTSKEVICH V.A., BELOUSOV A.A., 
Microstructural indicators of meat aging and taint, 
XVII European Congress of Workers of Meat 
Industry, England: 142, (1971). 
[5]. BEEM R., PLEVA B., Microscopy of meat and 
raw animal materials, Moscow: 336, (1964). 
[6]. GOREGLYAD H.S., Veterinary and sanitary 
examination with the basics of processing 
technology of animal products, Saint-Petersburg: 
Koloc, 584, (1981). 
[7]. TINYAKOV G.G., Histology of meat animals, 
Moscow: Food Industry, 460, (1967). 
[8]. http://medbib.in.ua/opredelenie-stepeni-
svejesti.html, 1/2015. 
[9]. JAYANTA KUMAR BASU, DEBNATH 
BHATTACHARYYA, TAI-HOON KIM, Use of 
Artificial Neural Network in Pattern Recognition, 
International Journal of Software Engineering and 
its Applications, Vol. 4, 2, 23-34, (2010).  
[10]. DEVROYE L., GYORFI L., LUGOSI G. A., 
Probabilistic Theory of Pattern Recognition, New 
York: Springer-Verlag, (1996). 
[11]. YOUGUO PI, WENZHI LIAO, MINGYOU 
LIU, JIANPING LU, Pattern Recognition 
Techniques, Technology and Applications, Theory 
of Cognitive Pattern Recognition, Vienna: In-Tech, 
432-462, (2008). 
[12]. MESTETSKY L.M., Mathematical methods 
of pattern identification. Course of lectures, 
Moscow: Faculty CMC of MSU, Department of 
MMP, 85, (2004). 

 

http://medbib.in.ua/opredelenie-stepeni-