Review on image-based animals weight weighing

Zhao, Yuliang; Xiao, Qijun; Li, Jinhao; Tian, Kaixuan; Yang, Le; Shan, Peng; Lv, Xiaoyong; Li, Lianjiang; Zhan, Zhikun

doi:10.1016/j.compag.2023.108456

Cited by 5 publications

(2 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While the prediction of cattle body weight is intricately tied to various indicators such as body length, height, width, rump height, and rump width, individually calculating these indicators proves cumbersome (Dohmen et al, 2022 ; Zhao et al, 2023 ). Moreover, the automatic application of these calculations in breeding barns often leads to significant errors.…”

Section: Methodsmentioning

confidence: 99%

Intelligent weight prediction of cows based on semantic segmentation and back propagation neural network

Xu,

Mao,

Wang

et al. 2024

Front. Artif. Intell.

View full text Add to dashboard Cite

Accurate prediction of cattle weight is essential for enhancing the efficiency and sustainability of livestock management practices. However, conventional methods often involve labor-intensive procedures and lack instant and non-invasive solutions. This study proposed an intelligent weight prediction approach for cows based on semantic segmentation and Back Propagation (BP) neural network. The proposed semantic segmentation method leveraged a hybrid model which combined ResNet-101-D with the Squeeze-and-Excitation (SE) attention mechanism to obtain precise morphological features from cow images. The body size parameters and physical measurements were then used for training the regression-based machine learning models to estimate the weight of individual cattle. The comparative analysis methods revealed that the BP neural network achieved the best results with an MAE of 13.11 pounds and an RMSE of 22.73 pounds. By eliminating the need for physical contact, this approach not only improves animal welfare but also mitigates potential risks. The work addresses the specific needs of welfare farming and aims to promote animal welfare and advance the field of precision agriculture.

show abstract

Section: Methodsmentioning

confidence: 99%

Intelligent weight prediction of cows based on semantic segmentation and back propagation neural network

Xu,

Mao,

Wang

et al. 2024

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Показана возможность измерения с помощью трехмерных камер массы тела и упитанности животных с вероятностью соответственно 0,89 и 0,63 [5]. Адекватный доступ к данным о весе является ключевым подходом к мониторингу здоровья скота и оценке производственных показателей [6].…”

Section: инновационные технологии и оборудование Innovative Technolog...unclassified

Performance of Cow Evaluation System Elements in Simulated Environmental Conditions

Yurochka,

Dovlatov,

Khakimov

et al. 2024

S-h. maš. tehnol.

View full text Add to dashboard Cite

Animal phenotype assessment plays a crucial role in enhancing the genetic potential of domestic breeds. Currently, research is underway to digitize and automate phenotyping through optical systems, enabling the recognition and evaluation of animals’ linear parameters. A prototype of a non-contact monitoring system for livestock phenotype indicators is currently under development. The movement of animals within the model (a laboratory measuring stand) is restricted by a specially designed transparent barrier made of polyethylene terephthalate, enabling the capture of three-dimensional photographs of the animals’ exterior. To validate the suitability of glass for this purpose, laboratory tests were conducted. (Research purpose) The research aims to determine the degree of relationship between protective glass contamination and the accuracy of the optical animal recognition system. (Materials and methods) The paper outlines the specifications of the laboratory stand, facilities, and equipment used, along with the methods employed and experimental conditions. The probability of determining the degree of protective glass contamination was quantified on a scale of 0 to 1, with a value of 0.78 indicating a high likelihood of accurate determination. (Results and discussion) The findings reveal that the system for determining the linear parameters of animals can operate reliably even when the protective glass is contaminated up to 30 percent. When the contamination reaches 50 percent due to inadequate glass cleaning, the system’s ability to recognize points of interest reduces by a factor of 1.625. Furthermore, at 80 percent contamination, achieving high-quality data collection becomes unfeasible as the camera fails to recognize the object. Proper cleaning of the glass is imperative to maintain the system stability. (Conclusions) The optical system enables the recognition and evaluation of animals’ linear parameters, provided that the protective glass of the laboratory stand is contaminated by no more than 50 percent and undergoes high-quality cleaning. At lower levels of glass contamination, up to 30 percent, these estimates exhibit a 2.6-38 percent increase compared to other contamination levels.

show abstract