ABSTRACT
Ground Penetrating Radar (GPR) is one of the most used devices for road structural damages detection. However, due to the different roadbed conditions and various disturbances in the nearby environment during detection, there are great difficulties in interpreting detection images, which also hinders automatic detection based on deep learning. In this work, we design a GPR image denoising method based on Cyclegan. We select the most suitable generator and add different attention mechanisms. After denoising the natural GPR road detection image, using the Yolo (You Only Look Once) to test the accuracy of the original image and the denoised image after adding different attention mechanisms. The detection accuracy is improved by 30%. The results of the detection network and the evaluation of the denoised images by GPR image interpreters indicate that the method has the following advantages: lower requirements for training data sets, a wide range of data sources, low cost, good denoising effect, and automatic detection of GPR images. It is of great help to the automatic detection of GPR images.
ABSTRACT
With the improvement of consumers' requirements for pork quality, the method of crossbreeding with excellent local pig breeds to improve meat quality is popular. Saba pig has high reproduction rate, good meat quality and high utilization rate of roughage, but its excellent characteristics have not been fully developed and utilized. To promote the development and utilization of Saba pigs and production of high-quality pork, the meat quality traits and glycolysis potential of Duroc × (Landrace × Yorkshire) (DLY), Berkshire × (Duroc × Saba) (BDS), and Duroc × (Berkshire × Saba) (DBS) three-way crossbred pigs were compared. The results showed that DLY had the highest live weight, carcass weight, lean meat percentage, drip loss, glycolysis potential, muscle diameter, and relative mRNA expression levels of type IIb muscle fibers as well as the lowest ultimate pH (p < 0.05). The lightness value of DBS was the highest (p < 0.05). Among the three crossbred pigs, myristic, arachidic, palmitoleic, and eicosenoic acids were the highest in BDS. These results indicated that the carcass traits of local crossbred pigs were worse than those of DLY pigs, but meat quality was markedly higher, with BDS showing the best meat quality.
ABSTRACT
OBJECTIVE: The intestinal microbiota plays an important role in host physiology, metabolism, immunity, and behavior. And host genetics could influence the gut microbiota of hybrid animals. The three-way cross model is commonly utilized in commercial pig production; however, the use of this model to analyse the gut microbial composition is rarely reported. METHODS: Two three-way hybrid pigs were selected, with Saba pigs as the starting maternal pig: Duroc× (Berkshire×Saba) (DBS) pig, Berkshire×(Duroc×Saba) (BDS) pig. One hundred pigs of each model were reared from 35 days (d) to 210 d. The body weight or feed consumption of all pigs were recorded and their feed/gain (F/G) ratio was calculated. On day 210, 10 pigs from each three-way cross were selected for slaughter, and cecal chyme samples were collected for 16S rRNA gene sequencing. RESULTS: The final body weight (FBW) and average daily gain (ADG) of DBS pigs were significantly higher than those of BDS pigs (p<0.05), while the F/G ratios of DBS pigs were significantly lower than those of BDS pigs (p<0.05). The dominant phyla in DBS and BDS pigs were Bacteroidetes (55.23% vs 59%, respectively) and Firmicutes (36.65% vs 34.86%, respectively) (p>0.05). At the genus level, the abundance of Prevotella, Roseburia, and Anaerovibrio in DBS pigs was significantly lower than in BDS pigs (p<0.01). The abundance of Eubacterium, Clostridium XI, Bacteroides, Methanomassiliicoccus, and Parabacteroides in DBS pigs was significantly higher than in BDS pigs (p<0.05). The FBWs and ADGs were positively correlated with Bacteroides, ClostridiumXI, and Parabacteroides but negatively correlated with the Prevotella, Prevotella/Bacteroides (P/B) ratio, Roseburia, and Anaerovibrio. CONCLUSION: These results indicated that host genetics affect the cecal microbiota composition and the porcine gut microbiota is associated with growth performance, thereby suggesting that gut microbiota composition may be a useful biomarker in porcine genetics and breeding.
