Showing posts with label computer vision. Show all posts
Showing posts with label computer vision. Show all posts

Saturday, 15 March 2025

A Device for Separating Silkworm Eggs by Color (Sex) Based on Computer Vision | Chapter 5 | Current Approaches in Engineering Research and Technology Vol. 10

Aims: In the production of high-quality hybrid silkworm eggs, it is very important to accurately divide the silkworm by sex so that the females of one breed are crossed with the males of another and vice versa. The practical implementation of this seemingly simple operation in production conditions is very difficult due to the lack of appropriate equipment and technology, and therefore it has not yet been possible to ensure a clear separation of silkworm eggs by sex. Therefore, the goal was to develop a device for separating silkworm eggs by color with software control based on computer vision.

Place and Duration of Study: Experiments were conducted at the Scientific Research Institute of Sericulture, Uzbekistan in the period 2019-2020 on eggs of silkworm breeds Belokokonnaya-1, Belokokonnaya-2, Saniish-8 and Saniish-9.

Methodology: The study and analysis showed that the separation of silkworm eggs is better carried out by piece-oriented feeding it to recognition zone, developed on the basis of computer vision using a high-speed web camera. It is also proposed to use Arduino UNO R3 as a hardware platform. The developed hardware and software for analysis and processing of data on the determination of color of eggs of silkworms of the developed device on a stationary platform is described. The device consists of four main blocks. The first one is designed for eggs feeding in portions, the second for the sequence of eggs alignment, the third is to bring these eggs to the estimation zone, where the fourth analyzes and separates light eggs from dark ones.

Results: When setting up the automation for the selection of light eggs, the separation error was 6-8%; with the dark silkworm eggs configuration, the error was in the range of 3 to 5%. A sufficiently high error when selecting light eggs is caused by the low contrast between the color of light silkworm eggs and the color of the transporting disk. So, at a high rotation speed, an egg merges with the background of the disk surface and webcam cannot always recognize it. Further, we plan to increase the separation accuracy by choosing a more precise illumination of eggs and an optimal rotation speed for the transporting disk. The recoverable losses of silkworm eggs were in the range of 1-2%, which corresponds to the required parameters.

Conclusion: The use of the device of such design will ensure reliability in operation, easiness of maintenance, high performance and accuracy, as well as eliminate silkworm egg's injuries and even its minimal losses. It will allow to introduction of new hybrids of the silkworm in production, and produce non-clogged, 100% hybrid silkworm eggs.

 

Author (s) Details

Mirzakhodjaev Bakhtiyar Anvarovich
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Mirzakhodjaev Anvar
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Bazarov Rustam Kamilyevich
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Tajenova Arzigul Tangatarovna
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Radzhabov Ismoil Bahadirovich
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Please see the book here:- https://doi.org/10.9734/bpi/caert/v10/3173

Tuesday, 14 January 2025

A Device for Separating Silkworm Eggs by Color (Sex) Based on Computer Vision | Chapter 5 | Current Approaches in Engineering Research and Technology Vol. 10

 

Aims: In the production of high-quality hybrid silkworm eggs, it is very important to accurately divide the silkworm by sex so that the females of one breed are crossed with the males of another and vice versa. The practical implementation of this seemingly simple operation in production conditions is very difficult due to the lack of appropriate equipment and technology, and therefore it has not yet been possible to ensure a clear separation of silkworm eggs by sex. Therefore, the goal was to develop a device for separating silkworm eggs by color with software control based on computer vision.

Place and Duration of Study: Experiments were conducted at the Scientific Research Institute of Sericulture, Uzbekistan in the period 2019-2020 on eggs of silkworm breeds Belokokonnaya-1, Belokokonnaya-2, Saniish-8 and Saniish-9.

Methodology: The study and analysis showed that the separation of silkworm eggs is better carried out by piece-oriented feeding it to recognition zone, developed on the basis of computer vision using a high-speed web camera. It is also proposed to use Arduino UNO R3 as a hardware platform. The developed hardware and software for analysis and processing of data on the determination of color of eggs of silkworms of the developed device on a stationary platform is described. The device consists of four main blocks. The first one is designed for eggs feeding in portions, the second for the sequence of eggs alignment, the third is to bring these eggs to the estimation zone, where the fourth analyzes and separates light eggs from dark ones.

Results: When setting up the automation for the selection of light eggs, the separation error was 6-8%; with the dark silkworm eggs configuration, the error was in the range of 3 to 5%. A sufficiently high error when selecting light eggs is caused by the low contrast between the color of light silkworm eggs and the color of the transporting disk. So, at a high rotation speed, an egg merges with the background of the disk surface and webcam cannot always recognize it. Further, we plan to increase the separation accuracy by choosing a more precise illumination of eggs and an optimal rotation speed for the transporting disk. The recoverable losses of silkworm eggs were in the range of 1-2%, which corresponds to the required parameters.

Conclusion: The use of the device of such design will ensure reliability in operation, easiness of maintenance, high performance and accuracy, as well as eliminate silkworm egg's injuries and even its minimal losses. It will allow to introduction of new hybrids of the silkworm in production, and produce non-clogged, 100% hybrid silkworm eggs.

 

Author(s)details:-

 

Mirzakhodjaev Bakhtiyar Anvarovich
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Mirzakhodjaev Anvar
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

Bazarov Rustam Kamilyevich
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Tajenova Arzigul Tangatarovna
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Radzhabov Ismoil Bahadirovich
Scientific Research Institute of Sericulture, Tashkent, Uzbekistan.

 

Please See the book here :- https://doi.org/10.9734/bpi/caert/v10/3173

Friday, 25 August 2023

Road Surface Image and Video Dataset for Machine Learning Applications with Seasons | Chapter 13 | Research Highlights in Science and Technology Vol. 9

Monitoring expressway surfaces is essential for ensuring the comfort and security of all road consumers, including boats and pedestrians. Furthermore, the maintenance of the roadways will benefit from this knowledge. As a result of the changeable weather, the state of the roads diminishes. Thus, producing an representation dataset of the road surface for two seasons-vacation and moist-thus serves as the main goal of the submitted paper. Consequently, we produced photos and videos of miscellaneous road surfaces, containing paved and unpaved roads. These folders have two subfolders for potholes in the moist and summer seasons. The dataset resides of 10 videos and 8484 pictures. For machine learning consultants working in the fields of automatic taxi control and road surface listening, this dataset is quite constructive.

Author(s) Details:

Sonali Bhutad,
Vishwakarma University, Pune, India.

Kailas Patil,
Vishwakarma University, Pune, India.

Please see the link here: https://stm.bookpi.org/RHST-V9/article/view/11681