Machine vision-based autonomous road hazard avoidance system for self-driving vehicles.

The resolution of traffic congestion and personal safety issues holds paramount importance for human's life. The ability of an autonomous driving system to navigate complex road conditions is crucial. Deep learning has greatly facilitated machine vision perception in autonomous driving. Aiming at the problem of small target detection in traditional YOLOv5s, this paper proposes an optimized target detection algorithm. The C3 module on the algorithm's backbone is upgraded to the CBAMC3 module, introducing a novel GELU activation function and EfficiCIoU loss function, which accelerate convergence on position loss lbox, confidence loss lobj, and classification loss lcls, enhance image learning capabilities and address the issue of inaccurate detection of small targets by improving the algorithm. Testing with a vehicle-mounted camera on a predefined route effectively identifies road vehicles and analyzes depth position information. The avoidance model, combined with Pure Pursuit and MPC control algorithms, exhibits more stable variations in vehicle speed, front-wheel steering angle, lateral acceleration, etc., compared to the non-optimized version. The robustness of the driving system's visual avoidance functionality is enhanced, further ameliorating congestion issues and ensuring personal safety.

The expression pattern of a rice proteinase inhibitor gene OsPI8-1 implies its role in plant development.

A rice proteinase inhibitor (PI) gene OsPI8-1 was identified. Belonging to the potato inhibitor I family, this gene contains a 201bp coding region with no introns and encodes a deduced protein of 66 amino acids which holds a PI domain. There are two uniform gene copies, OsPI8-1a and OsPI8-1b, with direct-repeat arrangement and an interval span of 13 kb on rice chromosome 8, corresponding to the site of BAC clone P0528B09 (Accession No. AP004703). Reverse transcription polymerase chain reaction (RT-PCR) assays showed that both OsPI8-1a and OsPI8-1b can be expressed in wild-type 'Zhonghua No.11'. To investigate the physiological functions of OsPI8-1 in plant development, we analyzed the expression patterns of the reporter gene beta-glucuronidase (GUS) driven by OsPI8-1 promoter at different developmental stages and tissues. It was demonstrated that no GUS signals were detected in the roots. Despite that very high GUS expression was examined in the shoot apical meristem, no detectable GUS activity in the developmental domains of leaf primordium was observed. OsPI8-1 promoter showed an obvious wound-induced response in mature leaves. Little GUS activity was detected in young nodes and internodes at the seedling stage, but active GUS expression was observed near the nodes on mature culms. In the developing stage of the anther, GUS signal was specifically located in the middle layer and the endothecium between the epidermis and tapetum. In the germinating seed, GUS expression was gradually accumulated in the side of scutellar epithelium close to the embryo. These tissue-specific accumulations suggested that OsPI8-1 has multiple endogenous roles on developmental regulation. In this report, the inhibitor function of OsPI8-1 to proteolytic enzymes and the potential influence of their poise on plant development (such as seed germination, tapetum degeneration, programmed cell death, etc.) were discussed.

Over-expression of rice OsAGO7 gene induces upward curling of the leaf blade that enhanced erect-leaf habit.

High-yield cultivars are characterized by erect leaf canopies that optimize photosynthesis and thus favor increased biomass. Upward curling of the leaf blade (called rolled leaf) can result in enhanced erect-leaf habit, increase erect duration and promote an overall erect leaf canopy. The rice mutant R05, induced through transferred DNA (T-DNA) insertion, had the rolled-leaf trait. The leaves in the wild type demonstrated natural drooping tendencies, resulting in decreasing leaf erection indices (LEIs) during senescence at the 20th day after flowering. Conversely, LEIs of the leaves in R05 remained high, even 20-day post-flowering. We applied T-DNA tagging and isolated a rolled-leaf gene from rice which, when over-expressed, could induce upward curling of the leaf blade. This gene encodes for a protein of 1,048 amino acids including the PAZ and PIWI conserved domains, belonging to the Argonaute (AGO) family. There are at least 18 members of the AGO family in rice. According to high-sequence conservation, the rolled-leaf gene in rice could be orthologous to the Arabidopsis ZIP/Ago7 gene, so we called it OsAGO7. These results provide a possible opportunity for implementing OsAGO7 gene in crop improvement.

[Genetic analysis of a heading-delayed mutant by T-DNA insertion in rice (Oryza sativa L.)].

Transposon tagging was used to isolate genes in higher plant. In this study, a delayed heading mutant caused by T-DNA insertion in rice was identified. Genetic analysis of the mutant showed that the three types of phenotype, normal heading, delayed heading and overly delayed heading in the segregating populations derived from the T-DNA heterozygotes fit the ratio of 1:2:1. Test for Basta resistance showed the delayed heading plants were all resistant while the normal heading plants were susceptible, and the ratio of resistant and susceptible plants was 3:1, which indicated that the delayed heading mutant was co-segregated with Basta resistance. The delayed heading mutant caused by T-DNA insertion was confirmed by T-DNA detection using PCR method. This delayed heading mutant will be used for isolation of the tagged gene in rice.

Distribution of T-DNA carrying a Ds element on rice chromosomes.

Over 3000 rice plants with T-DNA carrying a Ds element were constructed by Agrobacterium tumefaciens mediation. Using inverse PCR methodology, 590 unique right flanking sequences of T-DNA (Ds) were retrieved from independent transformants and classified into six main types on the basis of the origin of filler DNA between the right border of T-DNA and flanking sequence of rice genome. Type I sequences were the most common and showed canonical integration that T-DNA right border was followed by rice genome sequence with or without filler DNA of no more than 50 bp, while type II sequences displayed a vector-genome combination that T-DNA right border was followed by a vector fragment and then connected with rice genome sequence. The location and distribution of 340 type I and II flanking sequences on the rice chromosome were determined using BLAST analysis. The 340 Ds insertions at an average interval of 0.8 megabase (Mb) constructed a basic framework of Ds starter points on whole rice chromosomes. The frequency of T-DNA (Ds) inserted into the exons of predicted genes on chromosome one was 21%. Knowledge of T-DNA (Ds) locations on chromosomes will prove to be a useful resource for isolating rice genes by Ds transposon tagging as these Ds insertions can be used as starting lines for further mutagenesis.