Case Study: Swine Farming Measurement System

Overview
Livestock farming plays an important role in agriculture development, specifically for animal husbandry. Often bred for their meat, milk, fiber, fur, and other by-products, it is a time-consuming task that required day-to-day care. Beginning with selective breeding for the best possible production of by-products, feeding, maintaining the living environment, hygiene maintenance, medical attention, etc. to finally being brought to the slaughterhouse.
While taking care of animals may sound easy, there are plenty of factors that come into play, such as genetic origin (breed), balanced nutrition, hygiene, environment temperature, stress, etc. that may affect the animal’s growth in the long run.

Pain Points
While most factors are controllable, such as selecting a genetically-superior breed to maximize by-product production; monitor and controlling nutrition to ensure development within the growth curve; and environment conditioning via HVAC-like systems to control temperature, humidity, etc. However, the body/ size measurement inevitably comes down to human labor, which may be a stressful experience for the animal, and dangerous for both the animal the farmer, should the animal struggle. Not to mention that measuring hundreds and thousands of pigs is extremely time-consuming.

Problem-solved
Modern industrial computers and application solution vendors now have automated monitoring and measuring systems designed for farms to offer quality management and productivity improvement. In this case, a pig’s automated measurement system was deployed and the pig’s body size measurements were taken at 70, 100, and 130 days old in ten designated pens where the machine vision system is set up.

With individual pigs lured into each pen with food and closed off, the machine vision system scans for each pig’s body length, shoulder height/ width, hip height/ width. The identification of each pig can be done with electronic tags that can be detected and recognized. The scanned measurement data can then be uploaded via wireless communication.

Role of the Neousys Computer Platform
A general rundown of the hardware utilized include stepper motor controller and driver (SHUOKE M6505) to control the movement of the machine vision system, while the machine’s vision is achieved via binocular cameras (Basler acA-1600-20gc), which are connected via the Power over Ethernet port on the ultra-compact controller (Neousys POC-200). The serial ports on the Neousys POC-200 are also connected to the stepper motor controller and driver and digital input module (ART DAM-D3103).