Digital Dermatitis (DD)

Digital dermatitis (DD) is the most prevalent bovine infectious claw disease in North American and global cattle industries. The disease is responsible for painful circumscribed ulcerative lesions on the interdigital space of the hoof in both dairy and beef cattle. DD is associated with massive herd outbreaks of lameness and influences cattle welfare and production. Additionally, the disease results in major losses for the cattle industry because of severe lameness, decreased milk production, and increased infertility rate. Therefore, early detection can lead to prompt treatment, reduce costs, and decrease lameness within the herd.

Computer Vision

Traditionally, DD is determined through visual inspection or clinical experiments, but both methods require time and labor in addition to lacking the ability of early detection. Recent advances in computational power and algorithms have made object detection and classification possible through machine learning techniques. Computer vision can be used to perform object detection and calculate the associated class probabilities from a series of images or videos. Object detection locates the presence of an object in an image and draws a bounding box around that object. Such methods for animal detection have been used to assess the health of livestock in agriculture. However, applications for the classification of health events are still rare in veterinary medicine. Thus, computer vision provides a unique opportunity to improve early detection of DD in all cattle.

Some of the common model architectures used for object detection include variations of convolutional neural networks (CNNs) and region-based convolutional neural networks (R-CNNs). However, for the purposes of the study, You Only Look Once (YOLO) will be used since the methods are expected to be faster with a comparable or higher accuracy versus other computational approaches. The YOLO models will be used for real-time detection to predict bounding boxes and class probabilities from an image in a single evaluation. The study will compare and contrast the three latest versions of YOLO: YOLOv3, YOLOv4, and YOLOv5 for differences in performance.

You Only Look Once (YOLO)

You Only Look Once (YOLO) is one of the most powerful real-time object detector algorithms. It is called YOLO because unlike previous object detector algorithms, like R-CNN or its upgrade Faster R-CNN it only needs the image (or video) to pass one time through its network.

The YOLO models process 45 frames per second in real-time. YOLO views image detection as a regression problem from image pixels to bounding box coordinates and class probabilities. A single convolutional network simultaneously predicts multiple bounding boxes and class probabilities for those boxes and trains on full images and directly optimizes detection performance making its pipeline extremely fast and quite simple. It can process a streaming video in real-time with a latency of less than 25 seconds. During the training process, YOLO sees the entire image and is able to include the context in object detection.

In YOLO, each bounding box is predicted by features from the entire image. Each bounding box has 5 predictions: t_x, t_y, t_w, t_h, and confidence/objectness score p₀ where t_x, t_y is the centroid of the bounding box relative to the bounds of the grid cell and t_w, t_h are the predicted width and height of the whole image. This model is implemented as a convolutional neural network responsible for extracting the features, while the fully connected layers predict the coordinates and output probabilities.

Purpose

The general aim of the study is to minimizing effects of DD and lameness in all cattle via early detection and prompt treatments. The long-term goal will be to design, develop, and implement an application for the real-time detection of DD in dairy and beef cattle. Such tools have not yet been implemented and will help cattlemen improve DD prevention strategies for early intervention as well as helping increase cattle welfare and production.

Approach

1. Label images for M-stages of DD using the M-stage DD classification system images for DD in dairy and beef cattle (add to and rescore existing library of labeled images).
2. Augment and process library of labeled images to generate training dataset robust to oversampling and noise.
3. Train computer vision models: YOLOv3, YOLOv4, and YOLOv5.
4. Evaluate computer vision models for detection and scoring of DD at dairy and beef cattle farms using performance metrics.