Accurate estimation of body weight of pigs through smartphone image measurement app
The present investigation was carried out to develop a novel, portable and user-friendly method for hands-off measurement of body dimensions and prediction of body weight of the Ghoongroo breed of pigs (aged 30 to 190 days) using a smartphone object measurement app- On 3D CameraMeasure. Prediction o...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-08-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772375523000242 |
| _version_ | 1797838540220923904 |
|---|---|
| author | Gaganpreet Thapar Tapas Kumar Biswas Bharat Bhushan Syamal Naskar Amit Kumar Premanshu Dandapat Jaydip Rokhade |
| author_facet | Gaganpreet Thapar Tapas Kumar Biswas Bharat Bhushan Syamal Naskar Amit Kumar Premanshu Dandapat Jaydip Rokhade |
| author_sort | Gaganpreet Thapar |
| collection | DOAJ |
| description | The present investigation was carried out to develop a novel, portable and user-friendly method for hands-off measurement of body dimensions and prediction of body weight of the Ghoongroo breed of pigs (aged 30 to 190 days) using a smartphone object measurement app- On 3D CameraMeasure. Prediction of body weight from body dimensions was carried out through various types of regression analyses. This image-based method was compared with the traditional manual measurement method. Almost all the corresponding body dimensions between manual and image-based methods showed a very high and positive correlation (0.842–0.984). The product of body length and heart girth height measured from the images was identified as the best predictors of body weight. A comparison among stepwise regression and regression on one or two body dimensions showed that linear inclusion of more than one predictor did not increase accuracy. When non-linear regression using the product of two image-morphometric traits (Z=body length x heart girth height) was carried out, the R2 (0.98) value increased significantly. This sex and age-independent equation (Y= -6.335 + 0.024 Z, R2= 0.98) derived from the training dataset from an organized farm accurately predicted the body weights of pigs of an independent test population. The correlation coefficient between the actual body weight and the predicted body weight of pigs from the test population was 0.99. It was concluded that the smartphone image-based method could replace the cumbersome and stressful manual method very effectively. The additional advantages of smartphone image measurements are- robustness, accuracy, simplicity, user-friendliness, and true portability. |
| first_indexed | 2024-04-09T15:43:35Z |
| format | Article |
| id | doaj.art-e3faf98ce58f4d948e61bbbfe2a94990 |
| institution | Directory Open Access Journal |
| issn | 2772-3755 |
| language | English |
| last_indexed | 2024-04-09T15:43:35Z |
| publishDate | 2023-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj.art-e3faf98ce58f4d948e61bbbfe2a949902023-04-27T06:08:43ZengElsevierSmart Agricultural Technology2772-37552023-08-014100194Accurate estimation of body weight of pigs through smartphone image measurement appGaganpreet Thapar0Tapas Kumar Biswas1Bharat Bhushan2Syamal Naskar3Amit Kumar4Premanshu Dandapat5Jaydip Rokhade6Eastern Regional Station- Indian Veterinary Research Institute, ICAR, Kolkata 700037 WB, IndiaEastern Regional Station- Indian Veterinary Research Institute, ICAR, Kolkata 700037 WB, India; Corresponding author.Division of Animal Genetics and Breeding, ICAR-IVRI, Izatnagar, Bareilly 243122 UP, IndiaEastern Regional Station- Indian Veterinary Research Institute, ICAR, Kolkata 700037 WB, IndiaDivision of Animal Genetics and Breeding, ICAR-IVRI, Izatnagar, Bareilly 243122 UP, IndiaEastern Regional Station- Indian Veterinary Research Institute, ICAR, Kolkata 700037 WB, IndiaICAR- Central Avian Research Institute, Izatnagar, Bareilly 243122 UP, IndiaThe present investigation was carried out to develop a novel, portable and user-friendly method for hands-off measurement of body dimensions and prediction of body weight of the Ghoongroo breed of pigs (aged 30 to 190 days) using a smartphone object measurement app- On 3D CameraMeasure. Prediction of body weight from body dimensions was carried out through various types of regression analyses. This image-based method was compared with the traditional manual measurement method. Almost all the corresponding body dimensions between manual and image-based methods showed a very high and positive correlation (0.842–0.984). The product of body length and heart girth height measured from the images was identified as the best predictors of body weight. A comparison among stepwise regression and regression on one or two body dimensions showed that linear inclusion of more than one predictor did not increase accuracy. When non-linear regression using the product of two image-morphometric traits (Z=body length x heart girth height) was carried out, the R2 (0.98) value increased significantly. This sex and age-independent equation (Y= -6.335 + 0.024 Z, R2= 0.98) derived from the training dataset from an organized farm accurately predicted the body weights of pigs of an independent test population. The correlation coefficient between the actual body weight and the predicted body weight of pigs from the test population was 0.99. It was concluded that the smartphone image-based method could replace the cumbersome and stressful manual method very effectively. The additional advantages of smartphone image measurements are- robustness, accuracy, simplicity, user-friendliness, and true portability.http://www.sciencedirect.com/science/article/pii/S2772375523000242Smartphone image measurement appOn 3D camerameasureGhoongroo pigMorphometricsBody weight |
| spellingShingle | Gaganpreet Thapar Tapas Kumar Biswas Bharat Bhushan Syamal Naskar Amit Kumar Premanshu Dandapat Jaydip Rokhade Accurate estimation of body weight of pigs through smartphone image measurement app Smart Agricultural Technology Smartphone image measurement app On 3D camerameasure Ghoongroo pig Morphometrics Body weight |
| title | Accurate estimation of body weight of pigs through smartphone image measurement app |
| title_full | Accurate estimation of body weight of pigs through smartphone image measurement app |
| title_fullStr | Accurate estimation of body weight of pigs through smartphone image measurement app |
| title_full_unstemmed | Accurate estimation of body weight of pigs through smartphone image measurement app |
| title_short | Accurate estimation of body weight of pigs through smartphone image measurement app |
| title_sort | accurate estimation of body weight of pigs through smartphone image measurement app |
| topic | Smartphone image measurement app On 3D camerameasure Ghoongroo pig Morphometrics Body weight |
| url | http://www.sciencedirect.com/science/article/pii/S2772375523000242 |
| work_keys_str_mv | AT gaganpreetthapar accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp AT tapaskumarbiswas accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp AT bharatbhushan accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp AT syamalnaskar accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp AT amitkumar accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp AT premanshudandapat accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp AT jaydiprokhade accurateestimationofbodyweightofpigsthroughsmartphoneimagemeasurementapp |