Low Density Crops Research Articles

Mask R-CNN Refitting Strategy for Plant Counting and Sizing in UAV Imagery

This work introduces a method that combines remote sensing and deep learning into a framework that is tailored for accurate, reliable and efficient counting and sizing of plants in aerial images. The investigated task focuses on two low-density crops, potato and lettuce. This double objective of counting and sizing is achieved through the detection and segmentation of individual plants by fine-tuning an existing deep learning architecture called Mask R-CNN. This paper includes a thorough discussion on the optimal parametrisation to adapt the Mask R-CNN architecture to this novel task. As we examine the correlation of the Mask R-CNN performance to the annotation volume and granularity (coarse or refined) of remotely sensed images of plants, we conclude that transfer learning can be effectively used to reduce the required amount of labelled data. Indeed, a previously trained Mask R-CNN on a low-density crop can improve performances after training on new crops. Once trained for a given crop, the Mask R-CNN solution is shown to outperform a manually-tuned computer vision algorithm. Model performances are assessed using intuitive metrics such as Mean Average Precision (mAP) from Intersection over Union (IoU) of the masks for individual plant segmentation and Multiple Object Tracking Accuracy (MOTA) for detection. The presented model reaches an mAP of 0.418 for potato plants and 0.660 for lettuces for the individual plant segmentation task. In detection, we obtain a MOTA of 0.781 for potato plants and 0.918 for lettuces.

Strip Tillage and Sowing: is Precision Planting Indispensable in Silage Maize?

This work was aimed at assessing whether in silage maize it is possible to replace precisionplanting with a volumetric seeding in the perspective of developing hybrid machines to strip tilland sow both high density crops like winter cereals and low density crops like maize. This inorder to reduce the number of machines in the farm, simplify logistics and reduce amortizationcosts. Two experiments were carried out in 2014 and 2015. In the first year, two tillage-sowingtreatments were compared in a randomized block design with 5 replicates: 1) strip-tillage plusvolumetric band (0.1 m wide) seeding (ST-VBS) carried out by a Claydon Hybrid 6M at interaxle spacing of 0.6 m and with 35 kg ha-1 of seeds; 2) no-tillage plus precision line planting(NT-PLP) carried out by a sod drill Kinze 3100 at row distance of 0.71 m. In the second year,the same two treatments of 2014 were applied, but a third tillage-sowing treatment was alsoincluded: strip tillage plus precision line planting (ST-PLP) carried out by a strip tiller KhunStriger at inter axle spacing of 0.71 m plus the Kinze 3100, respectively, in two passages.In 2015, a randomized block design with 3 replicates was adopted. Both in 2014 and 2015treatments did not differ significantly for actual seeding density and final plant density,individual plant growth indices (plant height, stem diameter, FW, DW) at early stem elongation,flowering and final harvest, neither for total FW and DW yield, nor for biomass composition(starch, protein, lipids, fibre and ash concentrations) at harvest. Results demonstrate that a silagemaize crop can perform successfully when established by strip tillage associated withvolumetric band seeding. If similar results are demonstrated for high density crops, this willsupport the strategy of developing hybrid machines to strip till and sow both high density cropsand silage maize, which is relevant for many farming systems devoted to forage and biomassproduction for agro-energy purposes.

Machines for non-chemical intra-row weed control in narrow and wide-row crops: a review

Intra-row weed control in organic or low-input cropping systems is more difficult than in conventional agriculture. The various mechanical and thermal devices available for intra-row weed control are reported in this review. Low-tech mechanical devices such as cultivators, finger-weeders, brush weeders, and torsionweeders tend to be used in low density crops, while spring-tine harrows are mainly applied in narrow-row high-density crops. Flame weeding can be used for both narrow and wide-row sown crops, provided that the crop is heat-tolerant. Robotic weeders are the most recent addition to agricultural engineering, and only a few are available on the market. Nowadays, robotic weeders are not yet used in small and medium sized farms. In Europe, highincome niche crops are often cultivated in small farms and farmers cannot invest in high-tech solutions. Irrespectively of the choice of low- or high-tech machines, there are several weeders that can be used to reduce the use of herbicides, making of them a judicious use, or decide to avoid them.

Tiller dynamics and yield contribution in seed crops of Bromus inermis and Festuca pratensis established on different dates and with different plant densities

Tiller demography and contribution to seed yield were studied in first year seed crops of Bromus inermis ‘Løfar’ and Festuca pratensis ‘Salten’ planted on different dates and with increasing plant densities (A: 15 or 18 Jun/11 plants m−2, B: 15 or 30 July/44 plants m−2, C: 15 August or 10 September/178 plants m−2) in field trials at Landvik, SE Norway, in 1998 and 1999. While the total tiller population in most crops increased until seed harvest, it decreased during panicle elongation in crops of Bromus inermis and Festuca pratensis that had reached 2000 and 3500 tillers m−2 in early spring, respectively. Except for the fact that many of the primary tillers of Bromus inermis died after producing barren stems, tillers formed in August and September had the greatest chance of becoming reproductive and produced the longest stems with the heaviest inflorescences in both species. Most tillers produced in winter or early spring either remained vegetative or died, but spring-emerging tillers contributed up to 30% of the total seed yield in early-established, low-density crops of Festuca pratensis. The results are discussed in relation to former papers suggesting that inter-tiller transfer of flower induction stimuli is possible in some perennial grasses. It is concluded that tillers produced in winter and spring contribute more to seed yield in Festuca pratensis than in Bromus inermis.

SIMULATING RESPONSE OF MAIZE TO NITROGEN FERTILIZER IN SEMI-ARID ZIMBABWE

Data from a long-term trial on rates of fertilizer nitrogen (N) application to maize (Zea mays) were used to validate a cropping systems simulation model (APSIM) and then to apply the model to explore the risk associated with N fertilizer use by smallholder farmers and management strategies to minimize that risk. On average, maize growth and development in response to N was simulated with a degree of accuracy that justified its use in analysis of risk associated with N use in these semi-arid regions of Zimbabwe. APSIM was then configured to simulate the response to N over a 46-year climate record in order to assess the long-term risks associated with N use. The simulated long-term distribution indicated that negative responses to N could be expected in 15% of years, whereas no negative response to N was recorded in the experiments at the Makoholi Research Centre. Median responses were 20–30 kg maize grain kg−1 N for observed and simulated results. In terms of return on fertilizer investment, the observed and simulated distributions were also similar: in about 20% of years, a negative return could be expected, while in the best 20% of years a return of $Z5000 or more could be expected given the grain:fertilizer price ratio which, at March 2000, was about 1:7. The model analysis has suggested moderate rates (approximately 30 kg N ha−1) of N fertilizer would give greater responses per unit N applied than smaller rates (15 kg N ha−1). There was no evidence that conditional fertilizer strategies based on early-season rainfall would offer significant benefits over fixed application strategies. Early sowing at recommended population densities gave higher responses to N than were achieved for late sown or low-density crops.

O afilhamento em comunidades de cereais de estação fria é afetado pela qualidade da luz?

A radiação vermelha extrema refletida pelas folhas de plantas vizinhas altera a relação vermelho(V): vermelho extremo(Ve) e esta relação atua como um sinal precoce da competição que irá ocorrer em uma comunidade. O fitocromo é capaz. de detectar esta relação e transferir a excitação eletrônica em um sinal celular, que altera o metabolismo celular e influencia o crescimento e. o desenvolvimento da planta. Medições em comunidades em baixa densidade, onde ainda não há auto-sombreamento entre plântulas, demostram que a presença de plantas vizinhas altera de forma marcante a quantidade de Ve recebida pela superfície do caule. Nas regiões produtoras do Sul do Brasil, a pequena produção de afilhas férteis tem levantado uma séria questão sobre a importância do afilhamento para a formação do rendimento de grãos em cereais de estação fria. Neste sentido, as alterações na qualidade da luz no decorrer da formação da comunidade são importantes para a melhor compreensão deste processo. Portanto, a resposta à pergunta formulada no título depende da investigação dos efeitos da qualidade da luz sobre os cereais cultivados no Brasil, principalmente nos seguintes pontos: (i) os cereias em comunidade detectam precocemente as informações referentes ao nível de competição que irão sofrer? (ii) os cereais utilizam essas informações para modificar o padrão de afilhamento? (iii) há variabilidade genética entre os matérias cultivados no Brasil em relação a essas duas características ?

Borer infestation and damage in relation to maize stand density and water stress in the Ivory Coast

Abstract Borer infestations of maize crops grown at various planting densities were studied in the Ivory Coast. The number of larvae per stem for both stem borers, i.e. Busseola fusca (Fuller) (Lepidoptera, Noctuidae) and Eldana saccharina Walker (Lepidoptera, Pyralidae), was similar at different densities of planting in rainy conditions but was higher in low density crops in water stress conditions. The number of larvae of the cob borer Mussidia nigrivenella Ragonot (Lepidoptera, Pyralidae) per plant did not vary with crop density. Crop losses caused by borers decreased when the stand density was reduced and increased when the crop was suffering from water stress. The interactions between the various yield components and the compensation mechanisms that occurred, which were depending on the attack patterns, were studied.

Plant population studies on peanut (Arachis hypogaea L.) in subtropical Australia. 3. Growth and water use during a terminal drought stress

Two peanut cultivars of different botanical type (Virginia and Spanish) were grown at 3 plant population densities (40000, 120000 and 240000 plants/ha) and relied solely on stored soil water in a deep kraznozem soil for water requirements. Protracted crop water deficits occurred from flowering to maturity. Plant population influenced both the temporal and spatial patterns of water use, with high density crops extracting water from lower depths sooner than low density crops. Higher water use prior to early podfilling in high density crops was associated with more rapid leaf area development. Reproductive development was strongly influenced by plant population density, with more pods per m2 in low than in high density crops. Lower leaf water potential and individual leaf photosynthetic rates in the middle of the day during the pegging and early podding phase suggested that high crop water deficits had lowered assimilate availability and reduced reproductive potential in high, compared with low, density crops. The results indicate that there is scope for increasing pod yield when peanut is growing solely on stored water, by reducing plant population. The timing of water use, as distinct from the amount of water used, was the major determinant of pod yield.

Competition between annual weeds and vining peas grown at a range of population densities: effects on the crop

SummaryCompetition between annual weeds and vining peas (Pisum sativum L.) at five target population densities between 11 and 194 plants/m2 was examined by means of periodic destructive sampling of weedy and weed‐free plots. A further area of each plot was cut and vined to assess yields. Weeds impaired vegetative development, particularly by reducing tillering in low density crops. This resulted in weedy plots having fewer pods per plant at harvest but a lower proportion of flat pods than weed‐free plots. Weeds had no effect on numbers or weights of peas in full pods nor on tenderometer values of samples of vined peas.Adverse effects of weeds on the growth of individual crop plants decreased with increasing crop density. However, at lower crop densities many of the additional pods on weed‐free plots contributed little to vined yield, while at higher densities, direct or indirect effects of weeds increased the problem of maintaining sufficient photosynthetic area during pod swelling to prevent pod abscission and poor ovule development. Regression analysis of yield on crop density and of yield on numbers of pods per plant showed that vined yield per hectare was reduced by weeds by a constant amount across the range of densities and numbers of pods examined. Vining throughput was also reduced in weedy as compared with weed‐free crops, even on high density plots where little or no weed vegetation remained at harvest.In general, weed presence had effects similar to those caused by increasing crop density, but without the additional contribution to yield made by extra plants. Higher density crops suppressed weeds very effectively but were no less vulnerable to yield loss than those of lower density; they therefore merit just as much attention to effective weed control as crops suffering more visibly from competition by weeds.

Studies on carrot seed production: effects of plant density on yield and components of yield

SummaryThe yield of carrot seed (Chantenay types) increased by about 50%-55% with an increase in plant density from 10 to 80’plants m-2 in two experiments but in another there was no effect of density. The percentage of the yield contributed by the primary umbel was, on average, 25% and 62% for the low and high density crops, respectively. Increasing the plant density from 10 to 80 plants m-2 reduced the number of seeds per plant, largely because of fewer umbels, per plant. At anthesis, or soon after, seeds from primary umbels and low-density crops “were heavier, than those from secondary umbels and high-density crops. These differences were maintained throughout growth and, in . one year, were intensified by the shorter period of seed growth in the high-density crop. The lower mean seed weight in 1979 compared with 1980 was associated with a shorter period of seed growth and higher temperature after anthesis.

Seed quality in carrots: the effects of seed crop plant density, harvest date and seed grading on seed and seedling variability

SummaryCarrot seeds (Chantenay types) harvested 44–61 days after anthesis (early harvest) had higher coefficients of variation (CVs) of embryo length and subsequent seedling weight than those of seeds harvested 70–92 days after anthesis (late harvest). Seed grading after harvesting the seed crop reduced the CVs of some of the early harvested seed crops. Seed crops grown at high plant density (80 plants m—2) compared with low density (10 plants m-2) produced seed with lower CVs but the effects were small. Averaged over three experiments the CVs of seedling weight from early- and late-harvested seeds were 53% and 43%, respectively. For high- and low-density crops the corresponding figures were 49% and 47% and for ungraded and graded seeds 50% and 45%. The CV of seedling weight and the ‘spread’ of seedling emergence times were closely related to the CV of embryo length but not to the CV of seed weight. This suggests that, in addition to estimates of viability determined by standard tests, a measurement of the variability of embryo length would be a useful indication of subsequent crop variability.

The effects of ripening and drying of carrot seed (Daucus carota) crops on germination

SUMMARYCarrot seeds taken from the parent plant were capable of germinating before the stage when maximum seed dry weight was reached but even after this stage, when the seed moisture content had fallen below 20%, improvement in seed germination characteristics continued. In the latter stages of seed growth losses by shedding were 12–20 kg/ha/day. For low density crops (10 plants/m2) the yield of viable seed was at a maximum in crops whose seed was harvested with a moisture content of between 20 and 40% but no consistent relationship could be established for high density crops (80 plants/m2).There were no effects of umbel order or plant density on mean germination time or spread of germination. At any early harvest, percentage germination was highest for primary‐umbel‐seeds and seeds from low density crops but the differences between the seed origins diminished with later harvests.Drying the seeds on the umbels improved the percentage germination, reduced the mean germination time and the spread of germination particularly at the early harvests compared with seeds removed from the umbels and germinated immediately without drying.

The effects of seed size, autumn nitrogen and plant population density on the response to delayed sowing in winter oil-seed rape (Brassica napus)

SUMMARYThe aim of the experiments reported here was to investigate the possibility of minimizing yield losses due to late sowing by increasing leaf area and radiation interception during the pre-flowering period. The use of large seed did increase leaf area, radiation interception and crop weight at flowering. Normally this would be expected to lead to increased yield of late sowings, but hot, dry conditions severely affected all crops during seed growth. Applying fertilizer nitrogen in the seed bed boosted pre-flowering growth of early but not late sowings, probably because low temperatures prevented a response in the latter. Increasing plant population density did improve leaf cover and radiation interception until just before flowering, but leaf canopies in late-sown, low density crops were then able to expand more and function for longer owing to less shading by flowers and pods. Seed retention was improved and, even with as few as 8 plants/m3 in 1973–4, a worthwhile yield was still obtained.