University of Alberta professor co-develops AI imagery tech for dairy farmers
A new artificial intelligence automated imaging system is being brought into the farming industry to enhance the dairy cow classification process University of Alberta professor Reza Sabbagh, CEO of iClassifier and adjunct assistant professor at the University of Alberta, and Ghadar Manafiazar, an assistant professor in the Department of Animal Science and Aquaculture at Dalhousie University, have developed AI imagery tech for dairy farmers. The process of classifying cows without the use of technology, which can be costly, not readily available, and potentially potentially missing valuable indicators of health. They began working on automating the classification process by capturing images of dairy cows, processing them and using them to evaluate the animals. The new imaging system can scan up to 25 different traits, including bone quality and udder texture and depth, and is still being developed by the team. Both products are currently being developed with plans to launch once a year to assess a select number of cows.
Publicados : 2 anos atrás por Cindy Tran no Business Tech
We apologize, but this video has failed to load.
tap here to see other videos from our team. Try refreshing your browser, or Dairy farmers are currently classifying cows without the use of technology and the process is human-centred which, according to researchers, can be costly, not readily available and potentially missing valuable indicators of health. The process of classifying cows means farmers are looking at the structure of the animal, how they move and overall composition.
Reza Sabbagh, CEO of iClassifier and adjunct assistant professor at the University of Alberta, said his interest in developing a new technology came from several conversations he’s had with one of his partners on the project. What he heard was there were issues farmers were facing in assessing their animals. “There is an opportunity here, there is an issue farmers are (facing in) assessing the animals and they have some difficulties in doing that, and there’s opportunity that we can work on a product and a technology,” said Sabbagh.
“The issue is that right now, there is an assessment that is called classification for dairy animals for dairy cows, however, there are not enough classifiers to do the job at the time that producers need it. We thought if we can automate this process using AI, then we can help the producers because at the end of the day, this classification practice helped them to improve their productivity.”
• Local 'Godfather' of laser work inspired legacy shop as trend for tattoo removals rises Sabbagh, along with his co-founder, Ghadar Manafiazar, an assistant professor in the Department of Animal Science and Aquaculture at Dalhousie University, began doing the leg work on creating the technology.
They began trying to automate the classification process by working on capturing images of dairy cows, processing them and using them to evaluate the animals. Using an app, farmers can use their phone camera to scan an animal, or use a customized stall equipped with special imaging technology, which will process the images in the cloud using a unique AI algorithm. The imaging system can scan up to 25 different traits, including bone quality and udder texture and depth. Both products are still getting ready to be launched. “There are currently less than 20 registered classifiers, they travel across the country and see each individual animal and assess them. So then we said because of that limited resource, there’s a waiting time for a producer and every animal, if they are lucky to get chance of being assessed just once in their life,” said Manafiazar.
Dairy farmers like Nelson Jespersen, owner of Tuxedo Dairy Farms, said technology is entering the farming industry and he hopes the new AI imaging system will increase accuracy and productivity. He said right now he has someone coming once a year to classify a select number of cows. “It’s just the frequency of how we can classify and do lameness scoring. The biggest thing is that consistency, sometimes the human element isn’t as consistent and depending on who you get and what day you get them … I think this kind of technology creates better consistency and maybe accuracy, too, once it’s fully developed,” said Jespersen. Emran Hamzehnezhadabkenari, a master of engineering student at the University of Alberta, has been collecting data on the cows by monitoring cattle on a farm and detecting lameness in its early stages. “I’m doing data collection and I’m using all of these tools in order to do a health assessment, which is in an earlier stage,” said Hamzehnezhadabkenari. Manafiazar said in the future they hope they can enhance their technology to perform health assessments on all animals — but first they are perfecting their AI imaging system.
Tópicos: Academia, AI