LUMPY DISEASE

CLASSIFICATION USING
DEEP LEARNING
ABSTRACT
• Lumpy skin disease (LSD) is a severe and highly infectious pox disease of cattle caused
by the lumpy skin disease virus (LSDV).

• To facilitate early control of LSD, this study aimed to develop a new rapid on-site LSDV
detection method using an image processing tehnology to detect the lumpy diseases and
cattle monitoring.

• Now a day’s human cannot imagine their life without technology. Surrounding us diverse
technologies are helping people to live their lifestyles with more luxury.
INTRODUCTION
• We can say that the flow of animal lovers has gradually increased by the number of animal hospitals which proliferate
every year.

• There are so many benefits of having a pet including companionship guarding the house friendship as well as a part of a
family member.

• Pet owners now a days are very close to their pets and treat them as family members.

• That means health care or medical treatment is very crucial.

• One of the problems is caused by veterinarians or staff in health care industry is not being able to take care of the animals
24/7 (24 hours a day, 7 days a week)

• In general, a number of information is required to be able to identify the health of an animal.

• This information includes vital signs, pulse, breathing rate, temperature, blood-pressure, etc.

• This is used in a health surveillance system, as such data is a dataindication of life.

• If data abnormalities occur, that may show that the vital signs of animal may be dangerous.
LITERATURE SURVEY
Title Year Publication Author(s) Methodology Observation

A ConvNet based Real- CNN-based approach Achieved 95% accuracy in

International Conference N
using a dataset of hoof detecting lameness,
time Detection and 2022 on Communication, Shivaanivarsha; Pasupulet
images. Preprocessing demonstrating the
Computing and Internet i Baskaran Lakshmidevi; J
Interpretation of Bovine involved normalization effectiveness of CNNs in
of Things (IC3IoT) Teena Josy
and augmentation. automated detection.
Disorders

12th International Proposed a CNN Achieved 88% accuracy in

Cattle External Disease Conference on architecture for classifying respiratory
Classification Using Computing Md. Rony; Dola respiratory disease disorders. Real-time
2021
Deep Learning Communication and Barai; Riad; Zahid Hasan detection. Utilized real- processing demonstrated
Techniques Networking Technologies time video feeds from on- the feasibility for on-farm
(ICCCNT) farm cameras. applications.
LITERATURE SURVEY
Title Year Publication Author(s) Methodology Observation

The 3rd
FMD and Mastitis International Combined infrared Successfully
workshop on Recent Shivank Vyas, thermography with detected mastitis
Disease Detection CNNs for mastitis with 92% accuracy,
2019 advances on Internet nishant doshi, vipin detection. highlighting the
in Cows Using of Things: Shukla Integrated a two- complementary
Internet of Things Technology and stage approach for nature of thermal
Application improved accuracy. imaging and CNNs.
(IOT) Approaches

ResNet50
In Partial demonstrated
Identify Animal Compared different superior
Fulfillment of the
lumpy Skin Disease Requirements for CNN architectures performance with
for foot and mouth 94% accuracy,
Using Image 2021 the Degree of Elias Girma disease recognition. showcasing the
Master of Science in Explored transfer importance of
Processing and Computer Science learning techniques. selecting an
Machine Learning July appropriate
architecture.
MOTIVATION
• Animal Lumpy skin disease detection based on severity stages using image processing is an
untouched area.

• To reduce livestock mortality rate; due to skin diseases, early-stage detection is very important.

• The implementation of image processing technology in animal health industry has important
function by increasing the efficiency of experts.

• So, the motivation of this work is to contribute towards the achievement of animal skin disease
detection automatically in our country.
PROBLEM STATEMENT

• A Deep Learning Approach to Detect Lumpy Skin Disease in Cows. But Animal Lumpy skin
disease has different classification based on its stages.

• There is a need to further identify the different stages of Lumpy skin disease to know to what
extent the animal is affected by lumpy skin disease.

• Because the effective control Lumpy skin disease in endemic and non-endemic areas requires
rapid and accurate diagnostic methods to confirm a presumptive diagnosis .

• It is therefore the aim of this study to apply image processing and machine learning for detecting
the stages of lumpy skin disease in animals.
OBJECTIVES
• The general objective of this research is to design an appropriate detection model which
recognizes Lumpy animal skin disease using image processing techniques and machine learning
algorithms.

• The cattle monitoring using health parameter sensors and tracking.

• To prepare Animal Lumpy Skin Disease image data set for experimentation

• To select suitable segmentation and detection of Animal Lumpy Skin Disease.

• To construct detection model that identify Animal Lumpy Skin Disease with high sensitivity.

• To evaluate the performance of the proposed Lumpy Skin Disease detection model.
FUNCTIONAL REQUIREMENTS
• Image Input:
• The system should accept digital images as input for lumpy disease classification.
• Supported image formats should include common formats such as JPEG, PNG, etc.
• Preprocessing:
• Image preprocessing techniques should be applied to enhance the quality and extract relevant features.
• Techniques may include normalization, resizing, and data augmentation to increase the robustness of the model.
• Model Architecture:
• The CNN model should be designed specifically for lumpy disease classification.
• The architecture should include convolutional layers, pooling layers, and fully connected layers.
• The number of layers, filter sizes, and neurons in the fully connected layers should be defined based on the
complexity of the problem.
• Training:
• The system should support the training of the CNN using labeled datasets.
• Training should involve optimization algorithms (e.g., Adam), and the learning rate should be adjustable.
• The model should be able to converge within a reasonable number of epochs.
NON FUNCTIONAL REQUIREMENTS
• Performance:
• The system should achieve a specified level of accuracy in lumpy disease classification.
• Inference time should be within acceptable limits for real-time applications.
• Scalability:
• The system should be scalable to handle a growing dataset and increased computational demands.
• It should efficiently use resources as the dataset size expands.
• Robustness:
• The model should be robust to variations in input data, such as different imaging conditions and resolutions.
• It should handle noisy or incomplete data gracefully.
• Usability:
• The system should have a user-friendly interface for easy interaction.
• It should provide clear instructions for model usage, training, and evaluation.
• Security:
• Implement security measures to protect sensitive data used in the training and testing phases.
• Ensure that the deployed model is resistant to adversarial attacks.
• Reliability:
• The system should be reliable, minimizing the risk of false positives or false negatives.
• It should gracefully handle errors and exceptions.
• Maintainability:
• The codebase and documentation should be well-organized and easily maintainable.
• Provide updates and support for future improvements or changes in the model.
Components Required
• HARDWARE
• System : intel i3/i5 2.4 GHz.
• Hard Disk : 500 GB
• Ram : 4/8 GB
•
• SOFTWARE
• Operating system : Windows XP/ Windows 7.
• Software Tool : Open CV Python
• Coding Language : Python
• Toolbox : Image processing toolbox.

11
ARCHITECTURE
DATAFLOW DIAGRAM
Sequence diagram
Class diagram
Activity diagram
Use case diagram

User User
CONCLUSION
• In general, this study achieves better result towards detection of Lumpy skin disease and classify
as Sever, Mild and Normal skin.

• The contribution of this study includes preparation of Lumpy skin disease Image dataset,
construction of Lumpy skin disease Image classification model and Method to use local
information to known incidence of animal Epidemic disease.

• The main challenge observed in this study is the non-existence of Lumpy skin disease Image data
sate for experiment and Noises for properly detecting the region of interest.
 REFERENCES
• [1]. Abdi Feyisa, (2018), “clinical case studies on major diseases of veterinary importance in bishoftu town, Ethiopia”, unpublished

Master Thesis, College Of Veterinary Medicine And Agriculture Department Of Clinical Studies Addis Ababa university

• [2]. Teshome D, Derso S,(2015), “Prevalence of Major Skin Diseases in Ruminants and its Associated Risk Factors at University of

Gondar Veterinary Clinic, North West Ethiopia”. J Veterinar Sci Technol S: S13-002. Doi:10.4172/2157-7579.1000S13-002

• [3]. Ahmed Ali, (2018), “Review on lumpy skin disease and its economic impacts in Ethiopia”, Journal of Dairy, Veterinary & Animal

Research, Volume 7

• [4]. Zeedan GSG, Mahmoud AH, (2019), “Detection of lumpy skin disease virus in cattle using real-time polymerase chain reaction and

serological diagnostic assays in different governorates in Egypt in 2017”, Veterinary World, 12(7): 1093-1100.

• [5]. EFSA AHAW panel (EFSA Panel on Animal Health and Welfare ), (2015), “Scientific Opinion on lumpy skin disease”, EFSA

Journal 2015; 3986, 73pp.

• [6]. Betelihem Tegegne,(2018), “Outbreak investigation of lumpy skin disease; isolation and molecular characterization of the virus in

south wollo zone, northern Ethiopia”, unpublished master thesis, college of veterinary medicine and agriculture, Addis Ababa university