• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.8
• /
• pp.1560-1566
• /
• 2017

Recently, accumulation of data on pig farm is enabled through the wide spread of smart pig farm equipped with Internet-of-Things based sensors, and various machine learning algorithms are applied on the data in order to improve the productivity of pig farm. Herein, multiple machine learning schemes are used to predict the water usage in pig farm which is known to be one of the most important element in pig farm management. Especially, regression algorithms, which are linear regression, regression tree and AdaBoost regression, and classification algorithms which are logistic classification, decision tree and support vector machine, are applied to derive a prediction scheme which forecast the water usage based on the temperature and humidity of pig farm. Through performance evaluation, we find that the water usage can be predicted with high accuracy. The proposed scheme can be used to detect the malfunction of water system which prevents the death of pigs and reduces the loss of pig farm.

• Asian-Australasian Journal of Animal Sciences
• /
• v.8 no.4
• /
• pp.311-316
• /
• 1995

Data on available feed resources, feeding practices and nutrient adequacy of rations under small farm conditions in Sri Lanka were obtained in a baseline survey involving 104 pig farms. The results showed that a wide range of non-conventional feedstuffs are used for pig feeding under typical small farm conditions and that dietary protein quality is a major factor limiting productivity. Following the survey, two on-farm trials were conducted to evaluate cheaper, alternative feeding strategies. In trial 1, a test diet was formulated using several non-conventional feedstuffs and compared with a commercial feed that is normally fed in the farms. In trial 2, the possibility of improving growth rates by amino acid supplementation was evaluated. The results demonstrated that feed costs can be considerably lowered through these packages. Some problems inherent to on-farm livestock trials are highlighted.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.11
• /
• pp.457-464
• /
• 2019

The weight of pig is one of the main factors in determining the health and growth state of pigs, their shipment, the breeding environment, and the ration of feed, and thus measuring the pig's weight is an important issue in productivity perspective. In order to estimate the pig's weight by using the number of pig's pixels from images, acquired from a Top-view camera, the posture determining and the head removal from images are necessary to measure the accurate number of pixels. In this research, we propose the fast and accurate method to determine the pig's posture by using a fast image processing technique, find the head location by using a fast deep learning technique, and remove pig's head by using light weighted image processing technique. First, we determine the pig's posture by comparing the length from the center of the pig's body to the outline of the pig in the binary image. Then, we train the location of pig's head, body, and hip in images using YOLO(one of the fast deep learning based object detector), and then we obtain the location of pig's head and remove an outside area of head by using head location. Finally, we find the boundary of head and body by using Convex-hull, and we remove pig's head. In the Experiment result, we confirmed that the pig's posture was determined with an accuracy of 0.98 and a processing speed of 250.00fps, and the pig's head was removed with an accuracy of 0.96 and a processing speed of 48.97fps.

• Journal of Agricultural Extension & Community Development
• /
• v.16 no.4
• /
• pp.939-965
• /
• 2009

The Machine Learning has been identified as a promising approach to knowledge-based system development. This study aims to examine the ability of machine learning techniques for farmer's decision making and to develop the reference model for using pig farm data. We compared five machine learning techniques: logistic regression, decision tree, artificial neural network, k-nearest neighbor, and ensemble. All models are well performed to predict the sow's productivity in all parity, showing over 87.6% predictability. The model predictability of total litter size are highest at 91.3% in third parity and decreasing as parity increases. The ensemble is well performed to predict the sow's productivity. The neural network and logistic regression is excellent classifier for all parity. The decision tree and the k-nearest neighbor was not good classifier for all parity. Performance of models varies over models used, showing up to 104% difference in lift values. Artificial Neural network and ensemble models have resulted in highest lift values implying best performance among models.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.21 no.4
• /
• pp.203-210
• /
• 2001

To establish the recycling system of animal manure(AM) for environmental preservation and improve the utilization of AM, this study was to investigate the effects of the types and nitrogen application rate of AM on herbage productivity, efficiency of nitrogen utilization, nutritive value and an increase of soil fertility and in mixed grassland. This sudy was arranged in split plot design. Main plots were the types of AM(Cattle feedlot manure, CFM; Pig manure fermented with sawdust, PMFS; cattle sluny, CS) and subplots were the application rate of animal manure, such as 100, 200 and 300kgNiha. I. DM yields of herbage were the highest with CS and decreased by application over ZOOkgNiha AM. 2. Crude protein(CP) ontent was the highest with CFM and followed by CS, and the lowest with PMFS, and increased as application rate of AM increased. 3. Nitrogen(N) yields of CS treatment was higher than that of CFM and CS. and increased significantly as application rate of AM increased(P<0.05). 4. The contents of NDF, ADF and TDN was hardly influenced by the types and application rate of AM. 5. Organic matter(0M) content in the soil was the highest with PMFS and followed by CFM and the lowest with CS. OM content increased significantly as application rate of AM increased(P<0.05). 6. Total nitrogen content of the soil was not affected by the type of AM, but increased significantly as application rate of AM increased(P<0.05). (Key words : Animal manure, Grassland, Cattle feedlot manure, Pig manure fermented with sawdust, Cattle slurry, Soil fertility)

• Journal of Animal Environmental Science
• /
• v.5 no.3
• /
• pp.203-216
• /
• 1999

Recent public concern about air pollution caused by swine production facilities has forced to develop the methods to reduce and control the swine odors. Swine odors were affected the life of pig farm neighborhoods, swine productivity, pig health, diseases, and human right, safety, sanity as negatively. The first approaches of control of swine odors are the change or improve of the classical management systems, which are manure treatment method, manure storage facility, phase feeding, sex-divided feeding, feeder type, liquid-slurry feeding, environment control of swine building and dust control of indoor swine facility. The methods to control odor emission from manure have to include the diet modification as nutritional basis. In recent, research emphasis has focused on manipulating the swine diet to increase the nutrient utilization of the diet to reduce excretion products and reduction of odors. There are lots of feed additives and pit additives introduced as practical basis for reducing odor emissions. The ozone treatment method is candidate as the good system for reducing swine odor. But this system is still too expensive to practice in present.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.26 no.4
• /
• pp.293-300
• /
• 2006

This study was carried out to determine the effect of chemical fertilizer and two fermented types of pig slurry on the dry matter(DM) yield of three crops and chemical properties. The experiment design was a split plot with three replications. Main plots consist of three crops : $sorghum{\times}sudangrass$ hybrid('SS405'), sudangrass('Jumbo'), corn('DK501'). Sub plots consist of three treatments : chemical fertilizer (CF N-200, P-150, K-150 kg/ha), aerobic fermented pig slurry (APS 200kg N/ha), and aerobic fermented pig slurry treated with probiotics (APS+P 200 kg N/ha). Plant heights with three crops per sudangrass (380.3cm) was the longest (p<0.01). Dry matter yield of aerobic fermented pig slurry treated with probiotics was the highest the other treatments (p<0.01). Crude protein (CP) content were highest in $sorghum{\times}sudangrass$ hybrid than in the other crops. Cupper content(%) were highest in aerobic fermented pig slurry treated with probiotics than in the other treatments.

• Journal of Animal Environmental Science
• /
• v.8 no.1
• /
• pp.35-42
• /
• 2002

This study was carried out to determine the effect of chemical fertilizer and two fermented types of pig slurry on the total dry matter yield and quality of sorghum $\times$ sudangrass hybrid (Sorghum bicolor L. Moench) and chemical properties of soil after harvest in Jeju area. Three treatments consisted of chemical fertilizer (CF) 200kg 7kg Nha $^1$, aerobic fermented pig slurry (PS)200kg 7kg Nha $^1$, and aerobic fermented pig slurry treated with probiotics (PS+P) 200kg 7kg Nha $^1$were arranged in a randomized block design with three replications. The results obtained are summarized as follows. The heights of plant applied fermented pig slurry were slightly taller than those of plants applied chemical fertilizer during early growing stage, but there was no difference among treatments when the plants were harvested. The total forage dry matter yields were in the range of 14,848~ 15,42kg/ha, but they were not significantly different. Also, CP, NDF, ADF and mineral contents in the forage(% of DM basis) did not differ among treatments. The pH of soil was ranged from 5.35 to 5.63, but it was not significantly different. However, the content of available $P_2O_5$ of soil was higher(P<0.05) in chemical fertilizer treatment than that of soil in fermented pig slurry treatments. The content of K was higher(p<0.05) when ffrrmented pig slurry treated with probiotics was applied after the 1st harvest. but it was not different among treatments after the and harvest.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.3
• /
• pp.91-98
• /
• 2018

Automatic detection of pig wasting diseases is an important issue in the management of group-housed pigs. In particular, porcine respiratory diseases are one of the main causes of mortality among pigs and loss of productivity in intensive pig farming. In this paper, we propose a noise-robust system for the early detection and recognition of pig wasting diseases using sound data. In this method, first we convert one-dimensional sound signals to two-dimensional gray-level images by normalization, and extract texture images by means of dominant neighborhood structure technique. Lastly, the texture features are then used as inputs of convolutional neural networks as an early anomaly detector and a respiratory disease classifier. Our experimental results show that this new method can be used to detect pig wasting diseases both economically (low-cost sound sensor) and accurately (over 96% accuracy) even under noise-environmental conditions, either as a standalone solution or to complement known methods to obtain a more accurate solution.

• Journal of Biosystems Engineering
• /
• v.43 no.2
• /
• pp.119-127
• /
• 2018

Purpose: Fermented feedstuffs have been found to improve productivity, reduce manure odor, and increase immunity. However, because there is not a commercialized pelletizing system for fermented total mixed ration (TMR) for pig feeding in Korea, a pelletizing system using TMR fermented feed was developed. Methods: The particle size, density, and volumetric density of the TMR feeds used in the test were measured. The pellet durability index (PDI, %) value of the pelletized TMR feed based on its moisture content, and the amount of pellet production based on the rotation speed of the compression roller were measured. Results: The test materials, TMR1 and TMR2, were approximately compressed to 387 kg/m3 with 18.2% (w.b.) and 544 kg/m3 with 22.2% (w.b.), respectively. Throughout this pellet molding test, the moisture content from 15 to 20% (w.b.) of mixture feedstuffs, including fermented forage, could be used for pellet molding. Based on the results, a small-scale pellet molding system of fermented TMR was designed and manufactured for pig farms. As rotation speed increased, the throughput increased, whereas the moisture content decreased by approximately 2% (w.b.) because of pellet molding. The best yield of pellets with 94.2% PDI was of 536 kg/h at 135 rpm rotation speed. Conclusions: Although the throughput of the prototype increased as the rotation speed increased, it was difficult to operate because of the greater noise and the lower PDI (%) at the higher rotation speed of the pellet molding rotor. It was found that the best production of pellets using the prototype was 536 kg/h having a PDI of 94.2% or more at a rotation speed of 135 rpm.