Citation format:Ji Nan, Yin Yanling, Shen Weizheng, Kou Shengli, Dai Baisheng, Wang Guowei. Research Progress and Challenges of Crying in Pig Welfare Monitoring[J]. Smart Agriculture, 2022, 4(2): 19-35.
JI Nan, YIN Yanling, SHEN Weizheng, KOU Shengli, DAI Baisheng, WANG Guowei. Pig Sound Analysis: A Measure of Welfare[J]. Smart Agriculture, 2022, 4(2): 19-35.
Click to read directly to CNKI
Click to read directly to the official website (full text, free)
Research progress and challenges of crying in pig welfare monitoring
Ji Nan, Yin Yanling, Shen Weizheng*, Kou Shengli, Dai Baisheng, Wang Guowei
(College of Electrical Engineering and Information, Northeast Agricultural University, Harbin 150030, China)
Microphone-based audio technology is in a rapidly developing stage, and pig crying is one of the most important ways to assess pig welfare. Previous studies based on audio technology have focused on the relationship between specific pig sounds such as coughing, screaming or purring in pigs and pig welfare. At the same time, the existing research review focuses on the advantages and disadvantages of a variety of sensors commonly found in pig houses, and lacks the technical analysis and evaluation of specific sensor technologies in pig welfare monitoring. Therefore, this paper systematically expounds the existing research methods and evaluation indicators of pig cry as a means of welfare monitoring. In addition, from the technical level, each module of the pig cry audio analysis technology was summarized and analyzed. The challenges and prospects can be summarized as follows.
(1) It is difficult to obtain audio data at different growth stages. Although it is necessary to obtain sound data at different growth stages to analyze the complex and variable pig welfare, the current pig cry research is still in a specific growth stage. This situation is mainly due to the outbreak of African swine fever, which has led to stricter management of large commercial pig farms, and related research has had to be intermittent. This situation is likely to improve as the pandemic eases. At the same time, in the future, it can be considered to continue pilot trials in small and medium-sized pig farms.
(2) There is a lack of sound indicators and animal welfare monitoring and evaluation system. At present, the recognition and monitoring technology based on pig voice is still in the development stage of high technology dependence, and there is a lack of exploration and integration with animal welfare. Although it is very important to refine the details of pig farm experiments, including the analysis of sound in the house and the comparison of pig physiological indicators, there are no relevant studies in China. In order to overcome this dilemma, it is necessary to strengthen the participation and cooperation of researchers from different disciplines in the future, further explore the relationship between pig cry and disease, and construct monitoring standards and evaluation systems for this discipline.
(3) Individual pig welfare is difficult to monitor. At present, there are few studies on pig call localization and are still in the early stage of exploration. Although we can try to use multiple microphones in the experimental phase to improve the quality of sound data and the accuracy of target positioning calculations, in practical applications, we should consider achieving the best balance between the number of microphones, positioning accuracy, and cost. Microphone-based sound localization technologies are being upgraded, but they can only narrow the scope of sound monitoring as much as possible, and it is still difficult to achieve accurate analysis of individual pigs. In view of the importance of individual pig welfare monitoring, it will be an important research direction in the future to bring together the monitoring methods under different sensors, such as image monitoring, video monitoring and infrared monitoring, and combine the advantages of different methods to achieve multi-modal joint monitoring.
(4) Limited commercial product development. At present, most of the research on pig call recognition technology is in the algorithm research stage, and only foreign products such as SoundTalks and STREMODO system are still in the local regional trial stage. Microphone-based technology needs to widely consider the diversity and complexity of the pig farm environment, as well as the differences and dynamics of individual pigs, and other factors, therefore, there are still certain technical challenges in the transformation of commercial applications under the dual conditions of monitoring accuracy reliability and low equipment maintenance costs. At the same time, in view of the fact that pig farmers are still the core of precision animal husbandry at this stage, the recognition of new technologies by pig farmers should be fully considered. Although there is no industry consensus on precision animal husbandry, according to the survey, more than half of pig farmers expect to use new technologies to assist in pig raising, thereby reducing breeding costs and further improving the economic benefits of pig farms.
Finally, considering that the pig industry is in the stage of intensive and intelligent development, non-contact pig sound analysis is an important part of achieving precision animal husbandry, and it is undoubtedly one of the key technologies for the continuous progress of pig production in the future. Overcome technical limitations, deepen the integration and development of traditional pig animal husbandry industry and artificial intelligence technology, so as to effectively promote the research and development of related technical products, thereby accelerating the realization of the goal of precision animal husbandry, and promoting the high-quality and sustainable development of animal husbandry after transformation and upgrading.
Image of the article
Fig. 1 Flowchart of sound analysis
Note：Root mean square （RMS）；short-time energy （STE）；zero crossing rate （ZCR）；power spectral density （PSD）；Mel frequency cepstrum coefficients （MFCCs）；linear prediction cepstral coefficient （LPCCs）；short-time Fourier transform （STFT）
Fig. 2 Audio features used in analysis of pig sound
Source: Smart Agriculture, Issue 2, 2022
Please contact the editorial department for authorization for reprinting
Supported units in this issue
Jinglan Yunzhi Internet of Things Technology Co., Ltd
Zhejiang Zhenshan Technology Co., Ltd
Weichai Lovol Heavy Industry Co., Ltd
About the corresponding author
Prof. Weizheng Shen
Shen Weizheng (1977.11—), a scientist of intelligent breeding positions in the national dairy industry technology system, deputy director of the Northeast Key Laboratory of Smart Agriculture Technology of the Ministry of Agriculture and Rural Affairs, chief expert of the digital agriculture collaborative innovation and promotion system in Heilongjiang Province, the person in charge of the "Key Special Project of Intergovernmental International Science and Technology Innovation Cooperation" of the National Key R&D Program, the leader of computer science and technology discipline of Northeast Agricultural University, academic talent, and the youth editorial board member of "Smart Agriculture (Chinese and English)", He is the executive director of the Information Technology Subcommittee of the Chinese Association of Animal Husbandry and Veterinary Medicine, and a member of the Smart Agriculture Professional Committee of the Chinese Association of Automation. He has presided over 15 projects such as the National Key R&D Program, the Provincial Science and Technology Major, and the National Natural Science Foundation of China; He has won 1 second prize of Heilongjiang Provincial Science and Technology Progress Award, 2 third prizes of Natural Science and Technology Award, 1 first prize of Teaching Achievement Award, and 1 first prize of Animal Husbandry Science and Technology Achievement Award; More than 40 authorized invention/utility model patents and software copyrights; He has published more than 70 academic papers, including 42 papers included in SCI and EI; Editor-in-chief and deputy editor-in-chief of 4 textbooks.
WeChat communication service group
In order to facilitate the academic exchange of readers, authors and reviewers in the field of agricultural science, promote the development of smart agriculture, and better serve the readers, authors and reviewers, the editorial department has established a WeChat communication service group, where questions related to issues in the professional field and submissions can be consulted. How to join the group: Add the editor's WeChat 331760296, note: name, unit, research direction, the editor pulls you into the group, and the institutional marketing and advertising personnel do not disturb.
Release of Information
Introduction of scientific research team and recruitment information, publicity and promotion of academic conferences and related activities