TWI838908B - Flying unmanned vehicle laser pest control system based on safety considerations - Google Patents
Flying unmanned vehicle laser pest control system based on safety considerations Download PDFInfo
- Publication number
- TWI838908B TWI838908B TW111138436A TW111138436A TWI838908B TW I838908 B TWI838908 B TW I838908B TW 111138436 A TW111138436 A TW 111138436A TW 111138436 A TW111138436 A TW 111138436A TW I838908 B TWI838908 B TW I838908B
- Authority
- TW
- Taiwan
- Prior art keywords
- image
- laser
- pest
- processing unit
- target
- Prior art date
Links
- 241000607479 Yersinia pestis Species 0.000 title claims abstract description 134
- 238000012545 processing Methods 0.000 claims abstract description 56
- 238000004458 analytical method Methods 0.000 claims abstract description 46
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 claims abstract description 20
- 230000005540 biological transmission Effects 0.000 claims description 16
- 238000013527 convolutional neural network Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000033001 locomotion Effects 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 241000238631 Hexapoda Species 0.000 abstract description 31
- 241001465754 Metazoa Species 0.000 abstract description 10
- 230000009286 beneficial effect Effects 0.000 abstract description 6
- 230000007613 environmental effect Effects 0.000 abstract description 3
- 241000255925 Diptera Species 0.000 description 16
- 241000196324 Embryophyta Species 0.000 description 10
- 201000010099 disease Diseases 0.000 description 9
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 9
- 235000013399 edible fruits Nutrition 0.000 description 9
- 239000000575 pesticide Substances 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 7
- 238000000034 method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- 238000009313 farming Methods 0.000 description 4
- 235000013305 food Nutrition 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 239000002689 soil Substances 0.000 description 4
- 235000013311 vegetables Nutrition 0.000 description 4
- 241000255588 Tephritidae Species 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 239000003337 fertilizer Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000012010 growth Effects 0.000 description 3
- 230000000670 limiting effect Effects 0.000 description 3
- 235000015097 nutrients Nutrition 0.000 description 3
- 238000009329 organic farming Methods 0.000 description 3
- 239000005416 organic matter Substances 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 241000238876 Acari Species 0.000 description 2
- 241000254173 Coleoptera Species 0.000 description 2
- 244000017020 Ipomoea batatas Species 0.000 description 2
- 235000002678 Ipomoea batatas Nutrition 0.000 description 2
- 241000238814 Orthoptera Species 0.000 description 2
- 230000003042 antagnostic effect Effects 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 201000004792 malaria Diseases 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 244000052769 pathogen Species 0.000 description 2
- 241001136249 Agriotes lineatus Species 0.000 description 1
- 244000144730 Amygdalus persica Species 0.000 description 1
- 241001124076 Aphididae Species 0.000 description 1
- 241000894006 Bacteria Species 0.000 description 1
- 244000241257 Cucumis melo Species 0.000 description 1
- 235000015510 Cucumis melo subsp melo Nutrition 0.000 description 1
- 241000254171 Curculionidae Species 0.000 description 1
- 244000019459 Cynara cardunculus Species 0.000 description 1
- 235000019106 Cynara scolymus Nutrition 0.000 description 1
- 241000086071 Euproctis fraterna Species 0.000 description 1
- 241000282412 Homo Species 0.000 description 1
- 241000257303 Hymenoptera Species 0.000 description 1
- 241000256602 Isoptera Species 0.000 description 1
- 241000002163 Mesapamea fractilinea Species 0.000 description 1
- 241000256259 Noctuidae Species 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 235000006040 Prunus persica var persica Nutrition 0.000 description 1
- 244000061456 Solanum tuberosum Species 0.000 description 1
- 235000002595 Solanum tuberosum Nutrition 0.000 description 1
- 241000256247 Spodoptera exigua Species 0.000 description 1
- 241001464837 Viridiplantae Species 0.000 description 1
- 206010052428 Wound Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- FJJCIZWZNKZHII-UHFFFAOYSA-N [4,6-bis(cyanoamino)-1,3,5-triazin-2-yl]cyanamide Chemical compound N#CNC1=NC(NC#N)=NC(NC#N)=N1 FJJCIZWZNKZHII-UHFFFAOYSA-N 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 235000016520 artichoke thistle Nutrition 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001055 chewing effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003967 crop rotation Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000009189 diving Effects 0.000 description 1
- 235000013601 eggs Nutrition 0.000 description 1
- 210000002615 epidermis Anatomy 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000035558 fertility Effects 0.000 description 1
- 230000035784 germination Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 239000002917 insecticide Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 230000002147 killing effect Effects 0.000 description 1
- 238000002386 leaching Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000813 microbial effect Effects 0.000 description 1
- 244000005700 microbiome Species 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000006272 natural pesticide Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 239000002420 orchard Substances 0.000 description 1
- 230000017448 oviposition Effects 0.000 description 1
- 230000020477 pH reduction Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 108091008695 photoreceptors Proteins 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001850 reproductive effect Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000003612 virological effect Effects 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
Images
Abstract
Description
本發明係提供一種基於安全考量之飛行無人載具雷射除蟲系統,尤指一種可透過無人載具自動偵測害蟲之位置,並對應發出雷射,以去除害蟲或使其失能,藉以抑制害蟲之繁衍,並可進行穩態飛行,並可據以識別其他生物圖像並發出警示,使提升於自動去除害蟲時之安全性者。 The present invention provides a flying unmanned vehicle laser pest control system based on safety considerations, particularly a system that can automatically detect the location of pests through an unmanned vehicle and emit lasers accordingly to remove or disable the pests, thereby inhibiting the reproduction of pests, and can perform steady-state flight, and can identify other biological images and issue warnings, thereby improving safety during automatic pest removal.
按,二十世紀起化學肥料、化學農藥的發明使病蟲害減少,產量上升、農友收入增加,因此使用化學農藥的慣行農業被大量使用;然而隨著低成本的農藥使用量不斷增加,農藥帶來的問題也逐漸浮現,由於過度使用化學肥料、化學農藥來提高生產力,嚴重地污染了環境及危害到人類的身體健康;長期使用大量化學藥物,會導致土壤酸、鹽化,更會破壞水資源及生態環境,嚴重污染人們的生活環境;此外,隨著噴藥次數增加,病蟲害會產生抗藥性,因此又需要增加藥劑濃度,此一結果將使病蟲抗藥性更加提升,而高濃度藥劑將會對環境造成更大的汙染,除了嚴重影響動植物,也將對人類食品安全構成威脅,因此如何發展友善環境農業刻不容緩;是以,營造友善環境農業,並降 低對環境之毒害,同時亦可有效防除病蟲害與雜草分食養分,現已成為重要的研究目標之一。 Since the 20th century, the invention of chemical fertilizers and pesticides has reduced pests and diseases, increased production, and increased farmers' incomes, so the customary use of chemical pesticides has been widely used in agriculture; however, with the increasing use of low-cost pesticides, the problems brought by pesticides have gradually emerged. Due to the excessive use of chemical fertilizers and pesticides to improve productivity, the environment has been seriously polluted and human health has been endangered; long-term use of large amounts of chemical pesticides will lead to soil acidification and salinization, and will also damage water resources and the ecological environment, seriously polluting people's living environment; in addition, as the number of spraying increases, pests and diseases will develop resistance, so the concentration of the pesticide needs to be increased. This result will further increase the resistance of pests and diseases, and high concentrations of pesticides will cause greater pollution to the environment. In addition to seriously affecting animals and plants, it will also pose a threat to human food safety. Therefore, how to develop environmentally friendly agriculture is urgent; therefore, creating environmentally friendly agriculture and reducing the harm to the environment, while also effectively preventing pests and diseases and weeds from sharing nutrients, has become one of the important research goals.
在友善的耕作環境中,有機農業為一種較不污染環境、不破壞生態,並能提供消費者健康與安全農產品的生產方式;有機農業之定義因各國法律之規定而不同,隨著農業技術的演變,有機農業法規的要求亦漸趨嚴格;根據我國農委會之定義,有機農業是遵守自然資源循環永續利用原則,不允許使用合成化學物質,強調水土資源保育與生態平衡之管理系統,並達到生產自然安全農產品目標之農業;並於2007年1月,農委會便開始實施「農產品生產及驗證管理法」,令有機農業及其產品納入法律規範。 In a friendly farming environment, organic agriculture is a production method that is less polluting and less damaging to the environment, and can provide consumers with healthy and safe agricultural products. The definition of organic agriculture varies according to the laws of each country. With the evolution of agricultural technology, the requirements of organic agriculture regulations have become increasingly stringent. According to the definition of Taiwan's Council of Agriculture, organic agriculture is an agriculture that abides by the principle of sustainable use of natural resources, does not allow the use of synthetic chemicals, emphasizes the management system of water and soil resource conservation and ecological balance, and achieves the goal of producing natural and safe agricultural products. In January 2007, the Council of Agriculture began to implement the "Agricultural Product Production and Certification Management Act", which included organic agriculture and its products in legal regulations.
根據行政院農業委員會農糧署於民國110年11月統計指出,有機栽培農戶數有4,413戶,而有機種植面積有11,722.3276公頃;根據行政院農業委員會的「農業經營現況」指出,109年我國農耕土地面積79.0萬公頃,全年農作物種植面積74.0萬公頃,108年底臺灣地區農牧戶數為77.5萬戶,其中有從事農牧業之戶數為71.8萬戶;農牧戶人口為269萬餘人;由此可知,有機種植面積只佔農耕土地面積的1.5%,有機栽培農戶數只佔有從事農牧業之戶數的0.6%。 According to statistics from the Department of Agriculture and Food under the Council of Agriculture, Executive Yuan in November 2011, there are 4,413 organic farming households, and the organic farming area is 11,722.3276 hectares. According to the "Current Status of Agricultural Management" of the Council of Agriculture, Executive Yuan, the area of arable land in Taiwan in 2010 is 790,000 hectares, and the annual crop planting is 2.36 million hectares. The cultivated area is 740,000 hectares. At the end of 2019, there were 775,000 farming and animal husbandry households in Taiwan, of which 718,000 were engaged in agriculture and animal husbandry. The farming and animal husbandry population was over 2.69 million. It can be seen that the organic planting area only accounts for 1.5% of the cultivated land area, and the number of organic farming households only accounts for 0.6% of the number of households engaged in agriculture and animal husbandry.
根據有機農業全球資訊網引述歐盟的報導指出,目前歐盟有機農業面積占總農業面積比率為8.5%,若依據近年成長趨勢預估,至2030年約可增加至15-15%,有機行動計畫則進一步以有機占比提升至25%為目標,因此我國於有機農業之發展遠落後歐盟,其主因在於,我國氣候炎熱多濕,沒有足夠低溫的冬天,使得蟲害病害綿續不絕不易防治,此外,也因為台灣處於高溫多濕之環境,有機質分解迅速,淋溶之激烈,養分的損失大,產生地力減退,導致國內土壤有機質含量普遍偏低,全國超過65%耕地土壤有機質含量在2%以下, 購買價昂的有機質肥料有其必需性,再加上土地承載負荷太大,平均每公頃要負擔24個人口,使得沒有足夠的土地實施輪作種植,因此先天的條件限制了有機農業的發展;財團法人農業科技研究院農業政策研究中心亦指出,在農產品項部分,台灣的有機稻米及蔬菜的技術成熟,但有機水果規模仍較小,其係因我國緯度關係,使易受氣候影響及多病蟲害,故作物生長期也較長,因此較少人投入栽種。 According to a report from the EU cited by the Organic Agriculture Global Information Network, the current organic agriculture area in the EU accounts for 8.5% of the total agricultural area. If the growth trend in recent years is estimated, it will increase to about 15-15% by 2030. The Organic Action Plan further aims to increase the organic ratio to 25%. Therefore, Taiwan lags far behind the EU in the development of organic agriculture. The main reason is that Taiwan's climate is hot and humid, and there is no sufficiently low temperature in winter, which makes pests and diseases difficult to control. In addition, because Taiwan is in a high temperature and humid environment, organic matter decomposes quickly, leaching is intense, and nutrient loss is large, resulting in a decline in soil fertility, leading to domestic soil The organic matter content is generally low. More than 65% of the country's arable land has an organic matter content of less than 2%. It is necessary to buy expensive organic fertilizers. In addition, the land load is too large. On average, each hectare has to support 24 people, which means there is not enough land for crop rotation. Therefore, the inherent conditions limit the development of organic agriculture. The Agricultural Policy Research Center of the Agricultural Science and Technology Research Institute also pointed out that in terms of agricultural products, Taiwan's organic rice and vegetable technology is mature, but the scale of organic fruit is still small. This is because of the latitude of our country, which makes it vulnerable to climate impact and many diseases and pests, so the crop growth period is also longer, so fewer people invest in planting.
我國病蟲害種類因作物種類繁多,害蟲對經濟農作物蔬菜、果樹與花卉等的主要危害可以分為:取食性危害、非取食性危害及傳播病害三種;其中,取食性危害害蟲能直接危害花卉的根、莖、葉、芽、果實、種子等部位,形成不同部位的破壞,取食性危害害蟲主要破壞有6種方式:咀食、卷葉或綴葉營巢、潛葉、鑽蛀、刺吸、成癭;非取食性危害在農作物的危害主要有2種:產卵傷害與鑽土危害;傳播病害則是害蟲危害造成傷口,為某些病原菌侵入的管道,許多種植物病毒性病害都是由害蟲傳播的;害蟲就其危害的部位可以分地下害蟲、食葉害蟲、刺吸害蟲及蛀食害蟲等幾類。不同部位上的重要害蟲種類及其為害特徵又可以分:為害根部或種球的有:蟋蟀類、蠐螬、根潛蠅、甘藷蟻象、黃條葉蚤、根蟎等種類,其取食根系破壞表皮,影響根部水分及養分的吸收,造成樹勢衰弱,嚴重時導致地上部整株枯萎;危害花卉球根、甘薯塊根、馬鈴薯塊莖部位的受害則影響發芽、商品價值及產量。危害莖、樹幹、枝條的害蟲種類多為:桃折心蟲、木蠹蛾、螟蛾類、天牛類、象鼻蟲類、白蟻、莖潛蠅、介殼蟲類等,其會導致枝條或莖被鑽蝕為害後,造成枝條折斷及新梢枯萎的現象,影響樹勢的生長;受害樹幹,有些會形成剝皮狀,嚴重時整株枯死;危害花部者,主要係金龜子類及甜菜夜蛾、斜紋夜蛾等,其將直接影響花卉的 商品價值,間接造成落花或果實的商品價值;危為害果實者,如:東方果實蠅、瓜實蠅成蟲產卵於果實內,幼蟲於果肉內鑽食,造成果實腐爛、落果等現象;而其中,果實具有較高的經濟價值,對蟲害的忍受性最低,只要遭受少數昆蟲為害,即造成嚴重的損失;另外,危害幼苗根部的害蟲有螻蛄、地老虎、金針蟲、蠐螬等,這些害蟲危害植物後,不僅造成減產,而且影響蔬菜的品質,降低商品價值;危害葉部的重要害蟲可以分為:蠂蛾類害蟲、蚜蟲類、薊馬類、蟎類、潛蠅類、粉蝨類等種類。 Due to the wide variety of crops in my country, the main damage caused by pests to economic crops such as vegetables, fruit trees and flowers can be divided into three types: feeding damage, non-feeding damage and transmission of diseases. Among them, feeding pests can directly harm the roots, stems, leaves, buds, fruits, seeds and other parts of flowers, causing damage to different parts. Feeding pests mainly have 6 ways of damage: chewing, rolling, Nesting on leaves or leaf bundles, leaf diving, leaf boring, piercing and sucking, and galling; there are two main types of non-feeding damage to crops: egg-laying damage and soil-boring damage; transmitted diseases are wounds caused by pests, which are channels for certain pathogens to invade. Many plant viral diseases are transmitted by pests; pests can be divided into several categories according to the parts they damage, such as underground pests, leaf-feeding pests, piercing and sucking pests, and boring pests. Important pests on different parts of the tree and their damage characteristics can be divided into the following categories: those that damage roots or bulbs include: crickets, grubs, root divers, sweet potato ants, yellow-striped leaf fleas, root mites, etc. They feed on the roots and damage the epidermis, affecting the absorption of water and nutrients by the roots, causing the tree to weaken, and in severe cases, causing the entire above-ground plant to wither; damage to flower bulbs, sweet potato tubers, and potato tubers affects germination, commercial value, and yield. The pests that harm stems, trunks, and branches are: peach heartworm, wood borer, borer, longhorn beetle, weevil, termite, stem miner, shell insect, etc., which will cause the branches or stems to be bored, resulting in branch breakage and new shoot withering, affecting the growth of the tree; some damaged trunks will become peeling, and in severe cases, the whole plant will die; those that harm the flowers are mainly beetles, beet armyworm, and lithotripsy, which will directly affect the commercial value of flowers and indirectly cause the commercial value of flowers or fruits to fall; those that harm fruits, such as oriental fruit flies and melon The adult fruit flies lay eggs in the fruit, and the larvae eat in the flesh, causing the fruit to rot and fall. Among them, fruits have a higher economic value and are the least tolerant to insect damage. As long as they are damaged by a few insects, they will cause serious losses. In addition, pests that harm the roots of seedlings include crickets, cutworms, wireworms, and grubs. After these pests harm plants, they not only cause a reduction in production, but also affect the quality of vegetables and reduce the commercial value. The important pests that harm the leaves can be divided into: moth pests, aphids, artichokes, mites, divers, powder flies, etc.
有機農業的病蟲害防治主要目的就是要採取各種非農藥的自然防治法,希望促使益蟲益菌能夠與害蟲病菌維持在良好的生態平衡狀態;其方法雖所在多有,但約可分為栽培防治、物理防治、生物防治與自然農藥防治;有機農業的目的是要適當保護有益昆蟲以及各種天敵動物和微生物族群,利用生物間相抗衡的功能牽制害蟲的繁衍,以達到自然生態平衡,使作物不受到嚴重傷害;在有機農業經營方式下,已經消失的益蟲和各種天敵動物、拮抗性微生物等都會逐漸再出現,帶來友善的耕作環境。 The main purpose of pest control in organic agriculture is to adopt various non-pesticide natural control methods, hoping to enable beneficial insects and bacteria to maintain a good ecological balance with pests and pathogens; although there are many methods, they can be roughly divided into cultivation control, physical control, biological control and natural pesticide control; the purpose of organic agriculture is to properly protect beneficial insects and various natural enemy animals and microbial communities, and use the antagonistic functions between organisms to restrain the reproduction of pests, so as to achieve natural ecological balance and prevent crops from being seriously damaged; under the organic agricultural management mode, the beneficial insects and various natural enemy animals, antagonistic microorganisms, etc. that have disappeared will gradually reappear, bringing a friendly farming environment.
澳洲Australian Centre for Field Robotics,曾發展以太陽能為主動力的田間機器人RIPPA,試圖用來做為除草與除蟲的機器人,其採用物理防治法除蟲,主要是用影像處理判斷害蟲位置,並透過控制吸塵器的吸口方式來吸走害蟲。在RIPPA田間機器人的展示影片中,除害蟲的部分是以藍色的寶特瓶蓋作為假想的害蟲,故其目前尚於實驗階段,應仍有技術上之問題須予克服;其中,其吸塵器口稍大,故有破壞栽種植物的隱憂,另外有些害蟲會吸附在葉菜上,因此要不破壞栽種植物,又要將害蟲吸走,其吸塵力的控制機制為重要之課題。 Australian Centre for Field Robotics has developed a solar-powered field robot called RIPPA, which it attempts to use as a weed and pest control robot. It uses physical pest control, mainly using image processing to determine the location of pests, and controls the suction port of the vacuum cleaner to suck away pests. In the demonstration video of the RIPPA field robot, the pest control part uses blue plastic bottle caps as imaginary pests, so it is still in the experimental stage and there should still be technical problems to be overcome; among them, the vacuum cleaner mouth is slightly larger, so there is a concern that it may damage the planted plants. In addition, some pests will attach to leafy vegetables, so in order to avoid damaging the planted plants and suck away the pests, the control mechanism of its suction force is an important topic.
而現有之研究中,亦具有精準雷射蟲害控制之文獻,其係結合單眼立體視覺(monocular stereo vision)、快速雷射掃描(rapid laser scanning)與智慧影像辨識(intelligent pest recognition)等關鍵技術,以應用於農業害蟲種群數量控制,其主要係實驗用於棗園果樹的害蟲防治上,結果指出於特定之光波長及功率下,二齡以上的台灣黃毒蛾幼蟲在照射1.2秒後,會使其無法進一步攝入食物,故其雷射病蟲害防治儀可以有效地控制害蟲種群的方法。 There are also literatures on precise laser pest control in existing research. It combines key technologies such as monocular stereo vision, rapid laser scanning and intelligent pest recognition to control the number of agricultural pest populations. It is mainly used in experiments to control pests of fruit trees in date orchards. The results show that under specific light wavelengths and powers, larvae of the two-year-old Taiwan yellow tussock moth will be unable to ingest further food after being irradiated for 1.2 seconds. Therefore, the laser pest control device can effectively control the pest population.
而於1988年即有文獻提出利用雷射去除蟲害,其係使用雷射照射密閉空間裡的果蠅,發現綠光比紅外光有較好的效果;2008年蓋茲基金會也贊助Intellectual Ventures實驗室雷射殺蚊計劃,藉以消除瘧疾,因殺蟲劑雖可有效殺蚊,但是對環境破壞非常嚴重,且蚊子會產生抗藥性;該計畫的目標是利用藍光DVD裡面的雷射加上感光器(CCD或CMOS),利用物理方法滅蚊,希望成本控制在兩千元新台幣以內,能幫助非洲地區對抗瘧疾;惟藍光DVD的雷射光能量太小,故Lighting Science團隊在2016發表成果於Optics Express,其工作原理首先發射低功率紅外線光,照射範圍可達30公尺,並放置反光板,利用紅外線攝影機監視這個反光板,一旦有昆蟲進入光網,會在反光板上留下剪影,而紅外線攝影機可根據量測到的剪影初步判定是不是蚊子,更可利用蚊子大小、翅膀拍打的頻率及飛行速度,用來判定是蚊子之性別;由於母蚊子比較大,翅膀拍打的頻率也比較低,一旦系統軟體判定是母蚊子,就會發射高能雷射將蚊子擊落;根據研究,雷射功率不需要高到完全燒掉蚊子的翅膀,只要能破壞蚊子飛行能力即可將蚊子擊落,惟其機器之體積龐大,且價格遠超乎預期,目故其目前仍於研究階段。 In 1988, there was a document proposing the use of lasers to remove pests. It used lasers to irradiate fruit flies in a closed space and found that green light was more effective than infrared light. In 2008, the Gates Foundation also sponsored the Intellectual Ventures laboratory laser mosquito control project to eliminate malaria. Although insecticides can effectively kill mosquitoes, they are very harmful to the environment and mosquitoes can develop resistance to them. The goal of the project is to use the laser in the Blu-ray DVD plus a photoreceptor (CCD or CMOS) to kill mosquitoes using physical methods. It is hoped that the cost will be controlled within NT$2,000 and can help fight malaria in Africa. However, the laser energy of the Blu-ray DVD is too small, so the Lighting Science team published the results in Optics in 2016. Express, its working principle is to first emit low-power infrared light, the illumination range can reach 30 meters, and place a reflector, and use an infrared camera to monitor the reflector. Once an insect enters the optical network, it will leave a silhouette on the reflector. The infrared camera can preliminarily determine whether it is a mosquito based on the measured silhouette. It can also use the size of the mosquito, the frequency of wing flapping and the flying speed to determine the gender of the mosquito. Since female mosquitoes are larger and the frequency of wing flapping is also lower, once the system software determines that it is a female mosquito, it will emit a high-energy laser to shoot down the mosquito. According to research, the laser power does not need to be high enough to completely burn the mosquito's wings, as long as it can destroy the mosquito's flying ability, the mosquito can be shot down. However, the size of the machine is huge and the price is far beyond expectations, so it is still in the research stage.
而現今另提供一種概如我國專利公開第202022698號之「定位和除去昆蟲的系統及方法」專利案,其主要係依據影像偵測昆蟲之所在位置,並可予以標記之,並可搭載於一無人機上,並透過構件以機械或化學方式處置昆蟲;而其主要之技術在於,其係透過攝像器獲取空間影像,並基於該影像判斷該昆蟲在該空間中的所在位置,然而,透過單一攝像器並無法獲取空間中之位置,因單一攝像器所擷取的只是二維的影像,要取到三維空間位置至少需要二個攝像器,而透過其於另一實施例所述,透過二影像之時間差以擷取昆蟲在空間中之實際位置,然而,其配置係需令攝像器移動之位置為已知值,其所需之技術及運算力非常高,故成本將大幅增加,難以實際應用於無人機上;此外,其亦提及,攝像器與雷射或指示標記的指向裝置會被放在離彼此至少10公分遠的地方,惟此設置對於移動式除蟲裝置是有困難度的,且2008年蓋茲基金會也贊助Intellectual Ventures實驗室雷射殺蚊計劃,投影式之除蚊裝置是可行的,但對於自主移動式除蟲裝置的限制較高,難以有效實現之;再查,202022698一案說明,其處理器可在影像一單格內,含有昆蟲以及移動裝置的空間之影像所作的分析,控制移動裝置使其移動到昆蟲的附近,且前往昆蟲的方向,可以從一單格之內昆蟲在影像中的所在位置而被預估,處理器係使用單格中兩物體之間的像素距離,定時地判斷單格中移動裝置與昆蟲之間的角距,並基此計算出移動裝置需要移動的距離和方向;然而,如前述者,由於單一攝像裝置只能得到平面的資訊,因此除非假設昆蟲的大小已知,然後根據昆蟲影像在擷取影像裡所佔據的大小才能估測昆蟲的距離,否則只能獲得昆蟲的移動方向資訊,無法計算出昆蟲的距離;此外,其另外說明除去昆蟲的方法可包括判斷在昆蟲所在位置的附近是否有一生物(或是可被輔助裝置的動作傷害到的物體或材料)的方 法,其利用空間中的動作相對大小或是材料的形狀、顏色來做判斷。但是一般昆蟲相對於避免被傷害的生物大小相差非常大。當利用動作相對大小判斷時,植物隨風的飄動可能與生物的移動相似而難以區別。另外要在同一個擷取畫面要以形狀來同時辨識昆蟲與生物幾乎是不可能的工作。 Currently, there is another patent case similar to the "System and Method for Locating and Removing Insects" in Taiwan's Patent Publication No. 202022698. It mainly detects the location of insects based on images, can mark them, can be mounted on a drone, and dispose of insects mechanically or chemically through components; and its main technology is that it obtains spatial images through a camera and determines the location of the insect in the space based on the image. However, the position in space cannot be obtained through a single camera because a single camera only captures two-dimensional images. It is difficult to obtain three-dimensional space. At least two cameras are needed to capture the actual position of insects in space. According to another embodiment, the time difference between two images is used to capture the actual position of insects in space. However, the configuration requires that the position of the camera moving is a known value. The required technology and computing power are very high, so the cost will increase significantly and it is difficult to actually apply it to drones. In addition, it is also mentioned that the camera and the pointing device of the laser or indicator mark will be placed at least 10 cm away from each other. However, this setting is difficult for mobile pest control devices, and in 2008, the Gates Foundation also sponsored Intellectual Ventures Laboratory Laser Mosquito Killing Project, projection-type mosquito removal device is feasible, but the restrictions on autonomous mobile insect removal device are high, making it difficult to effectively implement it; further investigation, the 202022698 case explains that its processor can analyze the image of the space containing insects and mobile devices in a single frame of the image, control the mobile device to move to the vicinity of the insect, and the direction to the insect can be estimated from the location of the insect in the image within a single frame. The processor uses the pixel distance between the two objects in a single frame to regularly judge the angular distance between the mobile device and the insect in the single frame, and calculates based on this. Calculate the distance and direction that the mobile device needs to move; however, as mentioned above, since a single camera can only obtain planar information, unless the size of the insect is assumed to be known, the distance of the insect can be estimated based on the size of the insect image in the captured image. Otherwise, only the information of the insect's moving direction can be obtained, and the distance of the insect cannot be calculated; in addition, it is further explained that the method of removing insects may include a method of determining whether there is a creature (or an object or material that can be harmed by the action of the auxiliary device) near the location of the insect, which uses the relative size of the action in space or the shape and color of the material to make the judgment. However, the size of general insects relative to the creatures that are avoided from being harmed is very different. When using relative size of motion to judge, the movement of plants in the wind may be similar to the movement of organisms and difficult to distinguish. In addition, it is almost impossible to identify insects and organisms by shape in the same captured image.
有鑑於此,吾等發明人乃潛心進一步研究對於無人載具除蟲之動態穩定性控制,及其於運作時之安全性考量,並著手進行研發及改良,期以一較佳發明以解決上述問題,且在經過不斷試驗及修改後而有本發明之問世。 In view of this, we, the inventors, have devoted ourselves to further study the dynamic stability control of unmanned vehicle pest control and its safety considerations during operation, and have started to carry out research and development and improvement, hoping to find a better invention to solve the above problems. After continuous testing and modification, the present invention was born.
爰是,本發明之目的係為解決前述問題,為達致以上目的,吾等發明人提供一種基於安全考量之飛行無人載具雷射除蟲系統,其包含:一移動載具,其配置一飛行控制模組、一雷射發射元件、一影像擷取裝置、一攝影裝置、一目標距離感測器、一飛行狀態感測單元及一處理單元;該飛行控制模組係用以控制該移動載具之移動,而該飛行狀態感測單元係用以感測該移動載具之至少一飛行狀態參數;該雷射發射元件具有一焦距值;該處理單元係耦接於該飛行控制模組、該雷射發射元件、該影像擷取裝置、該攝影裝置、該目標距離感測器及該飛行狀態感測單元;以及一分析模組,其係連結於該處理單元;且該分析模組係建置有一害蟲識別模型;藉之,該影像擷取裝置係用以擷取一影像,該攝影裝置係用以擷取一第二影像,該處理單元係將該影像、該第二影像、該焦距值及所述飛行狀態參數傳送至該分析模組,該分析模組係透過該害蟲識別模型以於該影像及該第二影像中識別一目標,並藉由該目標距離感測器係據以量測其距離值,並對應產生一位置資訊,並依據該位置資訊、所述飛行 狀態參數、該距離值及該焦距值以界定一控制指令,並回傳至該處理單元,令該處理單元依據該控制指令以控制該飛行控制模組將該移動載具移動至對應於該位置資訊處,以令該雷射發射元件所發雷射之焦距對應於該位置資訊之位置者。 Therefore, the purpose of the present invention is to solve the above problems. To achieve the above purpose, we, the inventors, provide a flying unmanned vehicle laser pest control system based on safety considerations, which includes: a mobile vehicle, which is equipped with a flight control module, a laser emitting element, an image capture device, a camera device, a target distance sensor, a flight status sensing unit and a processing unit; the flight control module is used to control The flight state sensing unit is used to sense at least one flight state parameter of the mobile vehicle; the laser emitting element has a focal length value; the processing unit is coupled to the flight control module, the laser emitting element, the image capture device, the photographic device, the target distance sensor and the flight state sensing unit; and an analysis module is connected to the processing unit; and the analysis module The device is provided with a pest recognition model; the image capture device is used to capture an image, the camera device is used to capture a second image, the processing unit transmits the image, the second image, the focal length value and the flight state parameter to the analysis module, the analysis module recognizes a target in the image and the second image through the pest recognition model, and measures the target distance sensor based on the target distance. The distance value is measured, and a position information is generated accordingly, and a control instruction is defined according to the position information, the flight state parameter, the distance value and the focal length value, and is sent back to the processing unit, so that the processing unit controls the flight control module to move the mobile vehicle to a position corresponding to the position information according to the control instruction, so that the focal length of the laser emitted by the laser emitting element corresponds to the position of the position information.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,更包含一警示裝置,其係配置於該移動載具,並對應耦接於該處理單元;該分析模組係建置有一安全辨識模型,該安全辨識模型係用以辨識至少一安全標的,該處理單元係將該第二影像傳送至該分析模組,該分析模組係藉由該安全辨識模型以識別該第二影像中是否具有所述安全標的,並於辨識該第二影像中具有所述安全標的時,該處理單元係令該警示裝置發出警示者。 The above-mentioned flying unmanned vehicle laser pest control system based on safety considerations further includes a warning device, which is configured on the mobile vehicle and correspondingly coupled to the processing unit; the analysis module is configured with a safety recognition model, and the safety recognition model is used to recognize at least one safety target. The processing unit transmits the second image to the analysis module, and the analysis module uses the safety recognition model to identify whether the second image contains the safety target. When the second image is identified to contain the safety target, the processing unit causes the warning device to issue a warning.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,其中,該攝影裝置為廣角式攝影機,而該影像擷取裝置為長距式攝影機者。 According to the above-mentioned flying unmanned vehicle laser pest control system based on safety considerations, the camera device is a wide-angle camera, and the image capture device is a long-distance camera.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,其中,所述飛行狀態參數係包含該移動載具之經度、緯度、高度、俯仰角、偏航角、滾轉角、速度及角速度之至少其一者。 According to the above-mentioned flying unmanned vehicle laser pest control system based on safety considerations, the flight state parameter includes at least one of the longitude, latitude, altitude, pitch angle, yaw angle, roll angle, speed and angular velocity of the mobile vehicle.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,更包含一旋轉載台,其係設置於該移動載具,並對應承載該雷射發射元件及該影像擷取裝置;以及至少一驅動裝置,其係耦接於該處理單元,且對應連結並驅動該旋轉載台,該處理單元係藉由驅動所述驅動裝置,以令該雷射發射元件所發雷射之焦距對應於該位置資訊之位置者。 The above-mentioned flying unmanned vehicle laser pest control system based on safety considerations further includes a rotating platform, which is arranged on the mobile vehicle and correspondingly carries the laser emitting element and the image capturing device; and at least one driving device, which is coupled to the processing unit and correspondingly connects and drives the rotating platform. The processing unit drives the driving device to make the focal length of the laser emitted by the laser emitting element correspond to the position of the position information.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,更包含一遠端控制裝置,其係設有一傳輸模組及該分析模組,且該傳輸模組係連結於 該分析模組;而該處理單元更設有一通訊模組,該通訊模組係訊號連結於該傳輸模組,以令該處理單元連結並訊號傳輸於該分析模組者。 The above-mentioned flying unmanned vehicle laser pest control system based on safety considerations further includes a remote control device, which is provided with a transmission module and the analysis module, and the transmission module is connected to the analysis module; and the processing unit is further provided with a communication module, and the communication module is signal-connected to the transmission module to enable the processing unit to connect and transmit signals to the analysis module.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,其中,該害蟲識別模型於該影像中識別之目標為一害蟲之口部影像者。 According to the above-mentioned flying unmanned vehicle laser pest control system based on safety considerations, the target identified by the pest recognition model in the image is the image of the mouth of a pest.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,更包含一控制模塊,其係連結於該處理單元及該雷射發射元件;該害蟲識別模型係用以於該影像中識別該目標及其對應之種類資訊,該控制模塊係依據該害蟲識別模型識別該目標之種類資訊,以透過脈波寬度調變(Pulse-width modulation,PWM)調整該雷射發射元件之功率強度,並據以控制其所發雷射之時間及間隔者。 The above-mentioned flying unmanned vehicle laser pest control system based on safety considerations further includes a control module, which is connected to the processing unit and the laser emitting element; the pest recognition model is used to identify the target and its corresponding type information in the image, and the control module recognizes the type information of the target based on the pest recognition model to adjust the power intensity of the laser emitting element through pulse width modulation (PWM), and accordingly controls the time and interval of the laser emitted.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,其中,該雷射發射元件所發雷射之功率為5W以下者。 According to the above-mentioned flying unmanned vehicle laser pest control system based on safety considerations, the power of the laser emitted by the laser emitting element is less than 5W.
據上所述之基於安全考量之飛行無人載具雷射除蟲系統,其中,該害蟲識別模型係經深度神經網路(Deep Neural Networks,DNN)、深度置信網路(Deep Belief Networks,DBN)、卷積神經網路(Convolutional Neural Networks,CNN)、區域卷積神經網路(Regions with Convolutional Neural Networks,R-CNN)、快速地區域卷積神經網路(Faster R-CNN)或YOLO(You only look once)演算法,以訓練而得者。 According to the above-mentioned flying unmanned vehicle laser pest control system based on safety considerations, the pest recognition model is obtained by training with deep neural networks (DNN), deep belief networks (DBN), convolutional neural networks (CNN), regions with convolutional neural networks (R-CNN), fast regional convolutional neural networks (Faster R-CNN) or YOLO (You only look once) algorithm.
是由上述說明及設置,顯見本發明主要具有下列數項優點及功效,茲逐一詳述如下: From the above description and configuration, it is clear that the present invention mainly has the following advantages and effects, which are described in detail as follows:
1.本發明藉由飛行狀態感測單元之設置,藉可感測該移動載具之飛行狀態參數,且分析模組建置有害蟲識別模型,藉以於影像擷取裝置擷取影 像時,分析模組可於影像中識別一目標(如:害蟲)及其對應之位置資訊,位置資訊、飛行狀態參數、距離值及雷射發射元件之焦距值以界定一控制指令而界定控制指令,使處理單元可依據控制指令,而精確且穩定的控制該飛行控制模組將該移動載具移動至對應於該位置資訊處,並可令雷射發射元件朝對應於該位置資訊之位置發出雷射,藉以去除害蟲或使其失能,使可自動化的進行害蟲之去除,以利於適當牽制害蟲的繁衍,以防止受損,並有助於達致自然生態平衡,使促進相關農業之發展。 1. The present invention can sense the flight state parameters of the mobile vehicle by setting up a flight state sensing unit, and the analysis module builds a pest recognition model, so that when the image capture device captures an image, the analysis module can recognize a target (such as a pest) and its corresponding position information in the image. The position information, flight state parameters, distance value and focal length value of the laser emission element are used to define a control instruction, so that the control instruction is defined. The management unit can accurately and stably control the flight control module to move the mobile vehicle to the location corresponding to the location information according to the control instructions, and can make the laser emitting element emit lasers toward the location corresponding to the location information to remove pests or disable them, so that pest removal can be automated to properly control the reproduction of pests to prevent damage, and help achieve natural ecological balance to promote the development of related agriculture.
2.本發明透過遠端控制裝置之設置,並將分析模組配置於遠端控制裝置,使害蟲識別模型對於影像識別之運算可交由位於遠端之分析模組處理,藉可降低移動載具之硬體配置,使移動載具可予輕量化配置,以利其整體之重量可控制在2公斤以內,亦降低其電力之損耗,並可提升其運作之效能者。 2. The present invention uses a remote control device and configures the analysis module in the remote control device, so that the calculation of the pest recognition model for image recognition can be handed over to the analysis module located at the remote end for processing, thereby reducing the hardware configuration of the mobile vehicle, so that the mobile vehicle can be configured to be lightweight, so that its overall weight can be controlled within 2 kilograms, and its power consumption can be reduced, and its operating performance can be improved.
3.本發明更進一步係透過配置攝影裝置及警示裝置之設置,藉可進一步識別人類、益蟲或其他動物之圖像,並可據以發出警示,使環境生物可主動迴避,使避免受到雷射,藉以提升本發明於自動去除害蟲之安全性者。 3. The present invention is further configured with a camera and an alarm device to further identify images of humans, beneficial insects or other animals, and to issue an alarm accordingly, so that environmental organisms can actively avoid being exposed to lasers, thereby improving the safety of the present invention in automatically removing pests.
1:移動載具 1: Mobile vehicle
11:飛行控制模組 11: Flight control module
12:飛行狀態感測單元 12: Flight status sensing unit
2:處理單元 2: Processing unit
21:通訊模組 21: Communication module
22:控制模塊 22: Control module
23:警示裝置 23: Warning device
3:旋轉載台 3: Rotating stage
31:雷射發射元件 31: Laser emitting element
32:可變換焦距影像擷取裝置 32: Image capture device with variable focal length
32’:影像擷取裝置 32’: Image capture device
33:目標距離感測器 33: Target distance sensor
34:驅動裝置 34: Drive device
35:攝影裝置 35:Photographic equipment
4:分析模組 4:Analysis module
41:傳輸模組 41: Transmission module
5:遠端控制裝置 5: Remote control device
S001~S005:步驟 S001~S005: Steps
第1圖係本發明之系統架構示意圖。 Figure 1 is a schematic diagram of the system architecture of the present invention.
第2圖係本發明之流程圖。 Figure 2 is a flow chart of the present invention.
第3圖係本發明之立體示意圖。 Figure 3 is a three-dimensional schematic diagram of the present invention.
第4圖係本發明之局部剖視示意圖。 Figure 4 is a partial cross-sectional schematic diagram of the present invention.
第5圖係本發明之使用狀態示意圖。 Figure 5 is a schematic diagram of the present invention in use.
第6圖係本發明另一實施例之系統架構示意圖。 Figure 6 is a schematic diagram of the system architecture of another embodiment of the present invention.
第7圖係本發明另一實施例之立體示意圖。 Figure 7 is a three-dimensional schematic diagram of another embodiment of the present invention.
關於吾等發明人之技術手段,茲舉數種較佳實施例配合圖式於下文進行詳細說明,俾供鈞上深入了解並認同本發明。 Regarding the technical means of our inventors, several preferred embodiments are described in detail below with accompanying drawings, so that you can have a deeper understanding and recognize the present invention.
請先參閱第1圖至第3圖所示,本發明係提供一種基於安全考量之飛行無人載具雷射除蟲系統,其包含:一移動載具1,其在一實施例中係可為多旋翼之無人機,惟其僅係舉例說明,並不以此作為限定;該移動載具1係配置有一飛行控制模組11、一飛行狀態感測單元12及一處理單元2,且於一實施例中,移動載具1係配置有旋轉載台3,並且於該旋轉載台3上設置有一雷射發射元件31、一可變換焦距影像擷取裝置32及一目標距離感測器33;其中,飛行控制模組11,可知悉者,其係可為移動載具1之旋翼,藉以控制移動載具1之位置、升降、俯仰角、偏航角、滾轉角及速度,其控制方式係屬習知技術,故在此不予贅述;而該飛行狀態感測單元12係用以感測該移動載具1之至少一飛行狀態參數,如:移動載具1之經度、緯度、高度、俯仰角、偏航角、滾轉角、速度及角速度之至少其一者,故可知悉者,飛行狀態感測單元12係可被配置為對應之慣性感測器、高度計、GPS或是精確度更高的RTK GPS定位裝置之至少其一者;而就雷射發射元件31之設置而言,其係可對應的發出雷射,而雷射之波長越短則越具傷害性,故若以效率考量,係可將雷射之波長設置為400nm
以下,若以安全性考量,則可設置為450nm至550nm之間,使其波長較接近綠色,使較不會被綠色植物吸收;在一實施例中,亦可將波長設置為其他波長,藉以於害蟲係屬於綠色之顏色時,亦可予作用之;且在一實施例中,係可將雷射發射元件31所照射雷射之光點約為直徑2mm以下,而功率為5W以下,利於輕量型無人機的攜帶;令雷射發射元件31根據害蟲的屬性,利用脈衝調變調整是當的功率,照射於害蟲的要害部位短暫時間,於本實施例中係設定在2秒以內,增加無人機快速移動除害蟲的機動性;並藉以將其予以殺死,或使其失能,如:無法攝入食物,藉以降低其繁殖能力,惟前述之配置僅係舉例說明,並不以此作為限定。
Please refer to FIGS. 1 to 3. The present invention provides a flying unmanned vehicle laser pest control system based on safety considerations, which includes: a
而可變換焦距影像擷取裝置32之配置,由於本發明主要係用以拍攝並辨識害蟲,故其係可為攝影機,可為單一鏡頭做光學伸縮,也可以是多個單焦鏡頭組成影像擷取裝置,藉以進行影像之擷取,而由於害蟲部分具有保護色,故在一較佳之實施例中,其係可配置為熱影像儀或多光譜儀,亦可為前述攝影機、熱影像儀或多光譜儀之組合,藉使更利於對於害蟲特徵之提取。
As for the configuration of the variable focal length
而處理單元2則係耦接於該飛行控制模組11、該雷射發射元件31、該可變換焦距影像擷取裝置32及該飛行狀態感測單元12;而為使其可予輕量化配置,使移動載具1連同其整體搭載之裝置物件可控制於2公斤以內,以降低其能量耗損,故處理單元2係可為嵌入式系統之架構設置;以及一分析模組4,其係連結於該處理單元2;且該分析模組4係建置有一害蟲識別模型;而可知悉者,害蟲識別模型係用以檢測目標之影像中是否具有害蟲,而一般的目標檢測可以採用區域選擇、特徵提取或是分類器的方式來處理,而較佳者,係基於影像學習法則,如:深度學習,其係以類神經網路
為架構,是對資料進行表徵學習的演算法;而於本發明中,其係可經深度神經網路(Deep Neural Networks,DNN)、深度置信網路(Deep Belief Networks,DBN)、卷積神經網路(Convolutional Neural Networks,CNN)、區域卷積神經網路(Regions with Convolutional Neural Networks,R-CNN)、快速地區域卷積神經網路(Faster R-CNN)或YOLO(You only look once)演算法,以訓練而得者;而於本實施例中,係透過即時性較高的YOLO智慧型辨識法則進行,並透過足夠之害蟲圖片資訊經過YOLO的訓練,以形成其訓練模型,其即為該害蟲識別模型;而為利於害蟲識別模型之建置,在一實施例中,也可透過iNaturalist API以作為害蟲識別模型進行害蟲之識別,惟其僅係舉例說明,並不以此作為限定。
The
藉此,如第1圖至第5圖所示者,該可變換焦距影像擷取裝置32係用以擷取一影像,該目標距離感測器33係據以量測該可變換焦距影像擷取裝置32所擷取目標影像之一距離值,而該處理單元2係將該影像、該距離值及所述飛行狀態參數傳送至該分析模組4,令該分析模組4係透過該害蟲識別模型以於該影像中識別一目標(即影像中之害蟲圖像),並藉以依據可變換焦距影像擷取裝置32之內部參數,如:可變換焦距影像擷取裝置32之焦距、位置及角度等數據,以及移動載具1之位置,藉可於影像中運算求得該目標對應之位置資訊,而藉由目標之位置資訊與移動載具1之相對位置,可界定出移動載具1之飛行路徑,並透過結合移動載具1之飛行狀態參數,即可據以界定一控制指令,並回傳至該處理單元2,使處理單元2依據該控制指令以控制該飛行控制模組11將該移動載具1移動至對應於該位置資訊處,並令該雷射發射元件31朝對應於該位置資訊之位置發出雷射,藉以如前述者,透過雷射之照射,以去除害蟲或使其失能。
Thus, as shown in FIGS. 1 to 5, the variable focus
而如前述者,由於害蟲識別模型對於影像辨識須較大之運算量,而為令本發明可進行即時運算,並且降低移動載具1搭載處理單元2之硬體配置,使其可輕量化設置,故在一較佳之實施例中,係可於欲去除害蟲之區域內佈建一遠端控制裝置5,其係設有一傳輸模組41及該分析模組4,且該傳輸模組41係連結於該分析模組4;而該處理單元2更設有一通訊模組21,該通訊模組21係訊號連結於該傳輸模組41,以令該處理單元2連結並訊號傳輸於該分析模組4;藉此,由於處理單元2係搭載於移動載具1上,故其係可將影像及飛行狀態參數,透過通訊模組21傳輸至傳輸模組41,使分析模組4可予接收並進行前述透過害蟲識別模型進行目標之影像的辨識,並計算其控制指令,使移動載具1可專注於其飛行控制、影像擷取及對於飛行狀態參數之取得,藉此除可降低其硬體配置使其可輕量化外,並可降低其耗能,進而可有助於提升移動載具1之航行距離及範圍;而就通訊模組21與傳輸模組41間之通訊,在一實施例中,係可係透過行動通訊、無線網路、藍芽、LoRaWAN、NBIot或ZigBee進行無線連結,而在另一較佳之實施例中,亦可透過於傳輸模組41配置指向性天線,並可於其架設旋轉平台,以掃描式天線的方式進行連結,以提升其收訊能力,惟其僅係舉例說明,並不予限定之。
As mentioned above, since the pest identification model requires a large amount of computation for image recognition, in order to enable the present invention to perform real-time computation and reduce the hardware configuration of the
而在一較佳之實施例中,為利於對於可變換焦距影像擷取裝置32內部參數之擷取,使提升對於目標位置資訊擷取的精確性,並利於控制雷射發射元件31發射之方向,故較佳者,該移動載具1係可於其一端配置有旋轉載台3,本實施例係將旋轉載台3配置於移動載具1底端進行說明,惟並不以此作為限定;其中,旋轉載台3係對應承載該雷射發射元件31、目標距離感測器33及該可變換焦距影像擷取裝置32;以及至少一驅動裝置34,其係耦接於該處理單元2,
且對應連結並驅動該旋轉載台3,藉此,該處理單元2係可透過該可變換焦距影像擷取裝置32以追蹤該目標,並藉以配合所述飛行狀態參數,進行運算並驅動所述驅動裝置34,以令該雷射發射元件31朝對應於該位置資訊之位置發出雷射,並可藉由將驅動裝置34之傳動位置回傳至分析模組4,使可據以擷取可變換焦距影像擷取裝置32及雷射發射元件31之位置及角度資訊,以利分析模組4可精確的求得目標之位置資訊,並據以界定控制指令,且可知悉者,該控制指令係可包含對於驅動裝置34控制之參數,使可依據除蟲之環境,鎖定追蹤目標(即害蟲),控制移動載具1較佳之滯留位置,並透過驅動裝置34對於旋轉載台3之角度控制,使雷射發射元件31可確實的朝對應於該位置資訊之位置發出雷射。
In a preferred embodiment, in order to facilitate the acquisition of the internal parameters of the variable focal length
而就驅動裝置34之設置而言,在一實施例中,其係可為單一或多軸向設置,使可多角度的控制旋轉載台3,在一較佳之實施例中,旋轉載台3係可被配置為可於至少二軸之方向上旋轉運動,故在一實施例中,如第4圖所示者,驅動裝置34係可為馬達,並可設置為單數或複數個,以單獨或同時配置於該旋轉載台3之底端或側端,藉以對旋轉載台3進行角度之調整,使可如前述者,控制雷射發射元件31及可變換焦距影像擷取裝置32之角度;惟其僅係舉例說明,並不予限定之。
As for the arrangement of the driving
另就本案之安全性考量而言,其安全性之部分包含對於移動載具1在飛行狀態之穩定性,以及所發出之雷射對於周遭生物之相關危害;故本發明對於安全性之考量,使於進行害蟲去除作業時,可識別其他生物之影像,以防止其他生物進入移動載具1之飛行區域,或受到雷射照射而產生不利影響,故在一較佳之實施例中,如第6圖及第7圖所示者,基於前述,係可將可變換焦距影像擷取裝置32配置為影像擷取裝置32’,使其可具有,或者不具備可變換焦距之
功能,且進一步設置一攝影裝置35,其係分別配置於該移動載具1,並對應耦接於該處理單元2;該攝影裝置35係用以擷取一第二影像,且該分析模組4係建置有一安全辨識模型,該安全辨識模型係用以辨識至少一安全標的,而安全標的係可為人類、益蟲或其他動物之圖像,且同前所述,該處理單元2係將可該第二影像傳送至該分析模組4,該分析模組4係藉由該安全辨識模型以識別該第二影像中是否具有所述安全標的,並於辨識該第二影像中具有所述安全標的時,該處理單元2係可控制移動載具1遠離,或待命而不予令雷射發射元件31啟動而發射雷射,而在一實施例中,係可透過配置一警示裝置23,其係耦接於該處理單元2,且於第二影像中具有所述安全標的時,該處理單元2係令該警示裝置23,透過聲音或燈光發出警示,使環境生物可主動迴避,使避免受到雷射之影響,使提升本發明整體之安全性者。
In terms of the safety considerations of this case, the safety part includes the stability of the
且須特別說明的是,由於攝影裝置35係須大範圍的對周遭生物進行搜索,故該攝影裝置35較佳者,係可配置為廣角式攝影機,且在一實施例中,亦可更進一步的透過攝影裝置35進行廣角大範圍之目標影像搜尋,以利分析模組4可預先的於第二影像透過該害蟲識別模型以於第二影像中識別目標,再透過影像擷取裝置32’進行害蟲影像之擷取及位置之輔助定位及追蹤鎖定,故影像擷取裝置32’係可配置為長距式攝影機;此外,透過攝影裝置35及影像擷取裝置32’同步進行影像之擷取,亦可透過演算法(如:雙目測距法)據以獲得目標之空間座標資訊,藉以利於分析模組4可據以界定該控制指令,而於本實施例中,旋轉載台3係對應承載攝影裝置35、雷射發射元件31、目標距離感測器33及影像擷取裝置32’,藉以配合前述對於旋轉載台3之控制,以利影像擷取裝置32’進行目標之鎖定追蹤;惟其亦僅係舉例說明,並不予限定之。
It should be particularly noted that, since the
此外,因雷射本身仍具有焦點,在焦點的能量密度最大,故於一較佳之實施例中,係更進一步擷取雷射發射元件31之焦距值,使前述進行害蟲之去除時,係可透過將雷射發射元件31之焦距對準於害蟲,藉可降低雷射發射元件31之功率,並可於照射到周遭環境或生物時,因其焦點並非位於其位置上,故可降低對周遭環境或生物之危害;而可知悉者,該處理單元2之配置,係可耦接於該飛行控制模組11、該雷射發射元件31、該影像擷取裝置32’及該飛行狀態感測單元12;是以,該處理單元2係將該影像、該焦距值、該距離值及所述飛行狀態參數傳送至該分析模組4,使該分析模組4可進一步準確地透過該害蟲識別模型以於該影像中識別一目標及其對應之位置資訊,並可依據該位置資訊、所述飛行狀態參數、該距離值及該焦距值以界定一控制指令,並回傳至該處理單元2,令該處理單元2依據該控制指令以控制該飛行控制模組11將該移動載具1移動至對應於該位置資訊處,以令該雷射發射元件31所發雷射之焦距對應於該位置資訊之位置者。
In addition, since the laser itself still has a focus, the energy density at the focus is the highest. Therefore, in a preferred embodiment, the focal length value of the
進一步而言,在一具體之實施例中,所述飛行狀態參數係包含該移動載具1之經度、緯度、高度、俯仰角、偏航角、滾轉角、速度及角速度之至少其一者,故可據以求得可變換焦距影像擷取裝置32之姿態,而分析模組4則係可依據影像中所識別之目標,進行其前置處理、輪廓編碼、去除雜點、膨脹包覆破碎區塊、區塊分離、特徵提取及比對,配合前述可變換焦距影像擷取裝置32之相關內部參數之提取,藉以對可變換焦距影像擷取裝置32之姿態進行補償,藉可界定控制指令使移動載具1飛行至對應於該位置資訊處外,並可更進一步控制驅動裝置34驅動該旋轉載台3,使雷射發射元件31之焦點可確實地落在害蟲處;而較佳者,該害蟲識別模型於該影像中識別之目標為一害蟲之弱點影像,
如:口部或其翅膀之影像,而本實施例係透過將雷射發射元件31之焦點位於害蟲之弱點處,藉可使用較低功率或能量之雷射,舉例而言,係可令雷射發射元件31所發雷射之功率為5W以下,而雷射光點約為2mm以下,並予照射約1至2秒,使可據以去除害蟲或使其失能,亦可於不慎照射於環境或其他生物時造成相關危害,並可防止對農作物產生傷害
Furthermore, in a specific embodiment, the flight state parameters include at least one of the longitude, latitude, altitude, pitch angle, yaw angle, roll angle, speed and angular velocity of the
而由於害蟲之種類所在多有,為可因應害蟲之種類進行雷射功率之調整,故較佳者,係可更進一步配置一控制模塊22,其係連結於該處理單元2及該雷射發射元件31;該害蟲識別模型係用以於該影像中識別該目標及其對應之種類資訊,該控制模塊22係依據該害蟲識別模型識別該目標之種類資訊,以透過脈波寬度調變(Pulse-width modulation,PWM)調整該雷射發射元件31之功率強度,並據以控制其所發雷射之時間及間隔,使可更精確且有效地去除害蟲或使其失能。
Since there are many types of pests, in order to adjust the laser power according to the types of pests, it is better to further configure a
綜上所述,本發明所揭露之技術手段確能有效解決習知等問題,並達致預期之目的與功效,且申請前未見諸於刊物、未曾公開使用且具長遠進步性,誠屬專利法所稱之發明無誤,爰依法提出申請,懇祈 鈞上惠予詳審並賜准發明專利,至感德馨。 In summary, the technical means disclosed in this invention can effectively solve the problems of knowledge and achieve the expected purpose and effect. It has not been seen in publications before the application, has not been publicly used, and has long-term progress. It is truly an invention as defined in the Patent Law. Therefore, I have filed an application in accordance with the law and sincerely pray that the Supreme Court will give a detailed review and grant the invention patent. I will be very grateful.
惟以上所述者,僅為本發明之數種較佳實施例,當不能以此限定本發明實施之範圍,即大凡依本發明申請專利範圍及發明說明書內容所作之等效變化與修飾,皆應仍屬本發明專利涵蓋之範圍內。 However, the above are only several preferred embodiments of the present invention, and should not be used to limit the scope of implementation of the present invention. In other words, all equivalent changes and modifications made according to the scope of the patent application of the present invention and the content of the invention specification should still fall within the scope of the present invention patent.
1:移動載具 1: Mobile vehicle
11:飛行控制模組 11: Flight control module
12:飛行狀態感測單元 12: Flight status sensing unit
2:處理單元 2: Processing unit
21:通訊模組 21: Communication module
22:控制模塊 22: Control module
23:警示裝置 23: Warning device
31:雷射發射元件 31: Laser emitting element
32’:影像擷取裝置 32’: Image capture device
33:目標距離感測器 33: Target distance sensor
34:驅動裝置 34: Drive device
35:攝影裝置 35:Photographic equipment
4:分析模組 4:Analysis module
41:傳輸模組 41: Transmission module
5:遠端控制裝置 5: Remote control device
Claims (9)
Publications (2)
Publication Number | Publication Date |
---|---|
TWI838908B true TWI838908B (en) | 2024-04-11 |
TW202416074A TW202416074A (en) | 2024-04-16 |
Family
ID=
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190243385A1 (en) | 2016-10-18 | 2019-08-08 | Deakin University | Thrust vectored multicopters |
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190243385A1 (en) | 2016-10-18 | 2019-08-08 | Deakin University | Thrust vectored multicopters |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Williams et al. | Autonomous pollination of individual kiwifruit flowers: Toward a robotic kiwifruit pollinator | |
Bergerman et al. | Robotics in agriculture and forestry | |
CN107315381B (en) | Monitoring method for crop diseases and insect pests | |
Krishnan et al. | Robotics, IoT, and AI in the automation of agricultural industry: a review | |
US11526997B2 (en) | Targeting agricultural objects to apply units of treatment autonomously | |
US20160205917A1 (en) | Weed eradication method and apparatus | |
US11812681B2 (en) | Precision treatment of agricultural objects on a moving platform | |
CN105918281A (en) | Green and integrated insect killing method | |
KR20210010617A (en) | Artificial Intelligence Robot System for Unmanned Crop Cultivation and Harvesting | |
US20210186006A1 (en) | Autonomous agricultural treatment delivery | |
Bogue | Robots poised to transform agriculture | |
US11449976B2 (en) | Pixel projectile delivery system to replicate an image on a surface using pixel projectiles | |
CN102131392A (en) | Methods for increasing crop yield | |
TWI838908B (en) | Flying unmanned vehicle laser pest control system based on safety considerations | |
TW202416074A (en) | Flying unmanned vehicle laser pest control system based on safety considerations | |
TW202416073A (en) | Laser disinfestation method and system for intelligent dynamic and stable aiming unmanned vehicle | |
Rehna et al. | Impact of autonomous drone pollination in date palms | |
Tiwari et al. | Precision agriculture applications in horticulture. | |
Williams et al. | Evaluating the quality of kiwifruit pollinated with an autonomous robot. | |
Rajmane et al. | Precision agriculture and robotics | |
Masri et al. | Effectiveness of Bacillus thuringiensis aerial spraying against the bagworm, Metisa plana Walker (Lepidoptera: Psychidae) outbreak in oil palm using drone | |
KR20200124109A (en) | Artificial Intelligence Robot System for Unmanned Crop Cultivation and Harvesting | |
Grose | The next GREEN revolution | |
Fahrentrapp et al. | Autonomous UAV-based insect monitoring | |
Garg et al. | Revolutionizing Cotton Harvesting: Advancements and Implications |