Aerial Imagery Paddy Seedlings Inspection Using Deep Learning

In recent years complex food security issues caused by climatic changes, limitations in human labour, and increasing production costs require a strategic approach in addressing problems. The emergence of artificial intelligence due to the capability of recent advances in computing architectures could become a new alternative to existing solutions. Deep learning algorithms in computer vision for image classification and object detection can facilitate the agriculture industry, especially in paddy cultivation, to alleviate human efforts in laborious, burdensome, and repetitive tasks. Optimal planting density is a crucial factor for paddy cultivation as it will influence the quality and quantity of production. There have been several studies involving planting density using computer vision and remote sensing approaches. While most of the studies have shown promising results, they have disadvantages and show room for improvement. One of the disadvantages is that the studies aim to detect and count all the paddy seedlings to determine planting density. The defective paddy seedlings’ locations are not pointed out to help farmers during the sowing process. In this work we aimed to explore several deep convolutional neural networks (DCNN) models to determine which one performs the best for defective paddy seedling detection using aerial imagery. Thus, we evaluated the accuracy, robustness, and inference latency of one- and two-stage pretrained object detectors combined with state-of-the-art feature extractors such as EfficientNet, ResNet50, and MobilenetV2 as a backbone. We also investigated the effect of transfer learning with fine-tuning on the performance of the aforementioned pretrained models. Experimental results showed that our proposed methods were capable of detecting the defective paddy rice seedlings with the highest precision and an F1-Score of 0.83 and 0.77, respectively, using a one-stage pretrained object detector called EfficientDet-D1 EficientNet.

Adoption Status of Direct Seeded Rice Technology by the Farmers of Punjab

Direct-seeded rice (DSR) is a possible alternative to conventional puddled transplantedrice, where rice crop is seeded directly in non-puddled fields. The study was conducted toascertain the adoption status of recommended practices of DSR. Personal interviews wereconducted to collect data from randomly selected 210 farmers from three districtsrepresenting different agro-climatic regions of Punjab. Study revealed that selected farmershad adopted DSR on 49.20 per cent of their total area under paddy cultivation. Slightlyhigher number of selected farmers had adopted Tar-Wattar, a new technique of DSR overdry soil method. Only 18.52 per cent respondents had applied first irrigation as perrecommendations of Tar-Wattar method, whereas majority (67.59%) had applied firstirrigation before 21 days. 26.67 per cent farmers had adopted most suitable variety PR-126 and sown it at recommended time. However, in case of other varieties, majority offarmers did not follow recommended time of sowing. A smaller number of farmers haveapplied zinc, iron and sulphur fertilizers to ameliorate micro nutrient deficiencies. Averageyield of others varieties and PR 126 were found to be nearly identical, but due to lessercost of cultivation in DSR as compared to transplanted rice, adopter farmers’ fetched highernet return.

Respons Hasil Padi dan Hara Tanah Sawah terhadap Bakteri Pelarut Fosfat dan Pemfiksasi Nitrogen

The condition of unfertile paddy field due to excessive inorganic fertilizer application causing soil degradation. Fertilizing techniques using phosphate solubilizing and nitrogen fixer bacteria are breakthroughs that should be developed in paddy cultivation to increase fertility soil and harvest of rice. This research aims to study on phosphate solubilizing and nitrogen fixer bacteria ability to increase the harvest of rice by increasing chemical and biological fertility of soil. The research was conducted in the greenhouse of Sawah Baru experimental garden, IPB University. The design used was a completely randomized design with one factor that was dosage of inorganic fertilizers (NPK) with seven levels that were control, NPK 100%, NPK 0% + bacteria, NPK 25% + bacteria, NPK 50% + bacteria, NPK 75% + bacteria and NPK 100% + bacteria. Each stage was repeated four times, so there were 28 units. The results of phosphate solubilizing and nitrogen fixer bacteria could produce more rice grain 0.32 ton.ha^-1 than NPK 100% (recommended dose). Biological soil fertility improved with the increased population of bacteria, but there was no improved soil chemical fertility of soil.

Profile distribution of soil organic carbon fractions under different landforms in the Meghalaya plateau of India

Assessment of organic carbon fractions in soil provides the basis to ascertain vulnerability of an ecosystem to climate change. In the present study, we assessed SOC fractions in four pedons under contrasting landforms i.e., denudational low hill, upper plateau, lower plateau and valley in the Meghalaya plateau, India. Results indicated that soils of the studied pedons were acidic in nature, low in cation exchange capacity and base saturation. Further, surface (0-30 cm) soils were high in Walkley Black C (WBC) content (0.83-1.13%) in the studied pedons located under different landforms. The density of very labile carbon (VLC) fraction up to a depth of 150 cm was highest (49.22 Mg ha?1) in pedon 2 (P2) located in the upper plateau under shifting cultivation while that of less labile carbon (LLC) was highest (50.25 Mg ha?1) in pedon 4 (P4) in the valley under paddy cultivation. Higher densities of WBC and LLC in the valley (P4) as compared to other landforms in the study area indicate higher carbon sequestration potential of valley soil.

Distribusi Unsur Hara Kalium Tanah dan Kadarnya pada Tanaman Padi Sawah di Wilayah Sub Das Serayu Hilir Kecamatan Sampang Kabupaten Cilacap

This study aims to determine: (1) distribution of soil potassium nutrients in paddy fields in Serayu Watershed Downstream, Sampang District, Cilacap Regency, (2) distribution of soil potassium nutrient availability, K-content in plant tissue with rice yields in Serayu Watershed Downstream, Sampang District, Cilacap Regency, and 3) fertilizer recommendations potassium in paddy soil in Serayu Watershed Downstream, Sampang District, Cilacap Regency. The study method was conducted by determining 9 sample points by the making of SLH (Homogeneous Land Unit) map with a scale of 1:50.000 by overlaying the Sampang Disctrict Administration Map, the Soil Type Map, the Slope Map, and the Land Use Map. The variables mesured were pH H2O, pH KCl, DHL (electrical conductivity), redox potential, K-available, K-content in plant tissue, climatic conditions, plant varieties, and wet grain rice yields. The results showed that the distribution of potassium nutrients in Sampang District had a low status. The paddy varieties cultivated during research in Sampang District consist of Ciherang, HT Logawa, Inpari 32, and IR64. The climatic condition of Sampang District is the optimal climates for paddy cultivation. The best correlation result with crop yield was between K-available and crop yield. The value of K-available has a determinant coefficient of 47.37% of the yield of rice, while the K content in plant tissue has 19.45% of the yield of rice. Fertilizing recommendations in the research area is 87.73 kg K2O/ha or equivalent to an average of 175.47 kg KCl/ha.

Rancang Bangun Multi Sensor Pengukur Tinggi Muka Air Untuk Lahan Sawah

Abstrak. Penghematan air untuk irigasi permukaan adalah salah satu cara mengurangi efek lingkungan dari budidaya tanaman padi. Dalam sistem AWD(Alternate Wetting Drying) lahan sawah digenangi secara berselang, ketinggian air dari permukaan tanah dimonitor sesuai dengan rekomendasi AWD. Metode ini terbukti dapat mengurangi penggunaan air tanpa mengurangi hasil panen. Perkembangan saat ini pada platform cloud IOT sensor-sensor dapat berhubungan langsung dengan sistem yang menyimpan data dalam kapasitas yang besar, data ini selanjutnya dapat didownload dan diproses. Penelitian ini bertujuan untuk mendesain sistem multi sensor alat pengukur ketinggian permukaan air yang memiliki catu daya yang bersumber pada energi matahari dan dilegkapi dengan baterai Li-Ion dan berbentuk ringkas. Komunikasi dari masing masing sensor ke rangkaian master dilakukan melalui komunikasi radio. Mode hemat daya yang menonaktifkan sensor dan modul radio pada kondisi stanby dapat menghemat pemakaian baterai, tanpa penyinaran matahari baterai dapat bertahan 11,6 hari dibadingkan dengan 3,2 hari pada mode non hemat daya.Tegangan baterai dan pembacaan ketinggian permukaan air diupload ke ThinkSpeak IoT cloud oleh rangkaian master, untuk dianalisa lebih lanjut. Design of Paddy Field Water Level Multi Censor Abstract. Water conservation for field irrigation is an important key to reducing environmental effects from paddy cultivation. In the AWD (Alternate Wet Drying) method, the rice field is alternately flooded and non-flooded, water height is always measured to adjust with AWD recommendation. This method proved for reducing water usage without decreasing yield. With recent development from IoT Cloud platform, sensors can interface directly to cloud systems and store much data so it can be downloaded and processed later. In this study, electronic water height multiple sensors based on infrared technology is equipped with solar panel and Li-Ion battery storage designed with a compact form. Communication from sensors to master done by radio wave. Power saving method which turnoff sensors and radio power in the standby condition can conserve battery to stand the power without sunlight 11.6 days compared to 3.2 days in non-power-saving design. Battery voltage and water level reading from each sensor is uploaded to Think Speaks cloud IoT dashboard, this data can be analyzed later for other purposes.