Practices and Applications of Convolutional Neural Network-Based Computer Vision Systems in Animal Farming: A Review

Li, Guoming; Huang, Yanbo; Chen, Zhiqian; Chesser, Gary Daniel; Purswell, J.L.; Linhoss, John E.; Zhao, Yang

doi:10.3390/s21041492

Cited by 102 publications

(67 citation statements)

References 175 publications

(410 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One challenge in annotating the images recorded in the turkey flock was the quality of the raw data. Although the cameras had a high resolution, and therefore a high number of pixels in an image, a good sampling rate with 25 frames per second, and suitable installation (meaning an appropriate distance between the camera and surface of interest [ 32 ]), several issues were identified that made it difficult to detect injuries.…”

Section: Discussionmentioning

confidence: 99%

“…At this point, 132 images (20 images from 12 different recording days) were randomly selected, rated by the human observers, and tested for pixel-exact agreement. To examine the performance of the assessment and the segmentation, the intersection over union (IoU) was calculated whereby IoU serves as a standard performance measure for segmentation [ 31 ] and evaluates the deviation between ground truth and predicted areas [ 32 ].…”

Section: Animals Materials and Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Learn to Train: Improving Training Data for a Neural Network to Detect Pecking Injuries in Turkeys

Volkmann

Brünger

Stracke

et al. 2021

Animals

View full text Add to dashboard Cite

This study aimed to develop a camera-based system using artificial intelligence for automated detection of pecking injuries in turkeys. Videos were recorded and split into individual images for further processing. Using specifically developed software, the injuries visible on these images were marked by humans, and a neural network was trained with these annotations. Due to unacceptable agreement between the annotations of humans and the network, several work steps were initiated to improve the training data. First, a costly work step was used to create high-quality annotations (HQA) for which multiple observers evaluated already annotated injuries. Therefore, each labeled detection had to be validated by three observers before it was saved as “finished”, and for each image, all detections had to be verified three times. Then, a network was trained with these HQA to assist observers in annotating more data. Finally, the benefit of the work step generating HQA was tested, and it was shown that the value of the agreement between the annotations of humans and the network could be doubled. Although the system is not yet capable of ensuring adequate detection of pecking injuries, the study demonstrated the importance of such validation steps in order to obtain good training data.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Animals Materials and Methodsmentioning

confidence: 99%

Learn to Train: Improving Training Data for a Neural Network to Detect Pecking Injuries in Turkeys

Volkmann

Brünger

Stracke

et al. 2021

Animals

View full text Add to dashboard Cite

show abstract

“…Traditional image augmentation methods include rotating, flipping, scaling, cropping, etc. [ 22 ]. Because traditional data augment methods easily cause over fitting and under fitting, we present a GAN-based method for Martian rock image data generation, which can generate a large amount of true, diverse Mars images just by training with a few Martian rock images.…”

Section: Building Gmsrimentioning

confidence: 99%

GMSRI: A Texture-Based Martian Surface Rock Image Dataset

Zhang

Tian

et al. 2021

Sensors

View full text Add to dashboard Cite

CNN-based Martian rock image processing has attracted much attention in Mars missions lately, since it can help planetary rover autonomously recognize and collect high value science targets. However, due to the difficulty of Martian rock image acquisition, the accuracy of the processing model is affected. In this paper, we introduce a new dataset called “GMSRI” that is a mixture of real Mars images and synthetic counterparts which are generated by GAN. GMSRI aims to provide a set of Martian rock images sorted by the texture and spatial structure of rocks. This paper offers a detailed analysis of GMSRI in its current state: Five sub-trees with 28 leaf nodes and 30,000 images in total. We show that GMSRI is much larger in scale and diversity than the current same kinds of datasets. Constructing such a database is a challenging task, and we describe the data collection, selection and generation processes carefully in this paper. Moreover, we evaluate the effectiveness of the GMSRI by an image super-resolution task. We hope that the scale, diversity and hierarchical structure of GMSRI can offer opportunities to researchers in the Mars exploration community and beyond.

show abstract

“…Modern dairy farms continue to grow in herd size and technology adoption for maintaining or improving the production and labor efficiencies needed to feed the growing human population [ 1 ]. The U.S. is the largest dairy producer in the world with 9.39 million milking cows on farms, producing 101.25 million metric tons of milk in 2020 [ 2 ].…”

Section: Introductionmentioning

confidence: 99%

“…Early strategies for monitoring ingestive behavior relied on observation by technicians [ 8 ], to obtain precise individual information on a few animals but is costly and impractical for monitoring large herds [ 9 ]. Imaging methods combined with image processing algorithms or deep learning techniques may provide contactless and non-invasive measures and potentially automate the detection process [ 1 ]. However, high-quality images/videos are prerequisites for this, and appropriately recording the whole jaw movement process without occlusion within herds could be problematic.…”

Section: Introductionmentioning

confidence: 99%

Classifying Ingestive Behavior of Dairy Cows via Automatic Sound Recognition

Xiong

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

Determining ingestive behaviors of dairy cows is critical to evaluate their productivity and health status. The objectives of this research were to (1) develop the relationship between forage species/heights and sound characteristics of three different ingestive behaviors (bites, chews, and chew-bites); (2) comparatively evaluate three deep learning models and optimization strategies for classifying the three behaviors; and (3) examine the ability of deep learning modeling for classifying the three ingestive behaviors under various forage characteristics. The results show that the amplitude and duration of the bite, chew, and chew-bite sounds were mostly larger for tall forages (tall fescue and alfalfa) compared to their counterparts. The long short-term memory network using a filtered dataset with balanced duration and imbalanced audio files offered better performance than its counterparts. The best classification performance was over 0.93, and the best and poorest performance difference was 0.4–0.5 under different forage species and heights. In conclusion, the deep learning technique could classify the dairy cow ingestive behaviors but was unable to differentiate between them under some forage characteristics using acoustic signals. Thus, while the developed tool is useful to support precision dairy cow management, it requires further improvement.

show abstract

Practices and Applications of Convolutional Neural Network-Based Computer Vision Systems in Animal Farming: A Review

Cited by 102 publications

References 175 publications

Learn to Train: Improving Training Data for a Neural Network to Detect Pecking Injuries in Turkeys

Learn to Train: Improving Training Data for a Neural Network to Detect Pecking Injuries in Turkeys

GMSRI: A Texture-Based Martian Surface Rock Image Dataset

Classifying Ingestive Behavior of Dairy Cows via Automatic Sound Recognition

Contact Info

Product

Resources

About