《電子技術應用》
您所在的位置:首頁 > 通信與網(wǎng)絡 > 設計應用 > 復雜背景下小尺寸多角度人臉檢測方法研究
復雜背景下小尺寸多角度人臉檢測方法研究
網(wǎng)絡安全與數(shù)據(jù)治理
黃杰,劉芬
天津職業(yè)技術師范大學電子工程學院
摘要: 為了提升復雜背景下小尺寸人臉檢測精度,提出了一種人臉檢測方法GhostNet-MTCNN。在多任務級聯(lián)卷積神經(jīng)網(wǎng)絡(MTCNN)主干網(wǎng)絡上,將占用計算資源的普通卷積進行舍棄,利用GhostNet網(wǎng)絡中計算量更低的Ghost bottleneck模組替代卷積的作用,重新構建網(wǎng)絡特征提取功能,從而搭建一個新的模型。實驗結果表明,該方法可以有效平衡參數(shù)量和精度。在Easy、Medium、Hard三種驗證集上,與MTCNN相比在參數(shù)量僅增加0.62M的前提下精度分別提升了 5.6%、6.6%、7.8%,與MobileNetV3-MTCNN相比在參數(shù)量減少1.27M的同時精度又分別提升了1.6%、0.8%、0.5%。該研究能夠在復雜場景下提高模型對小尺寸、多角度人臉檢測精度,同時也能夠有效平衡參數(shù)量和檢測精度使其成為在邊緣設備部署中更優(yōu)的選擇。
中圖分類號:TP18文獻標識碼:ADOI:10.19358/j.issn.2097-1788.2024.04.008
引用格式:黃杰,劉芬.復雜背景下小尺寸多角度人臉檢測方法研究[J].網(wǎng)絡安全與數(shù)據(jù)治理,2024,43(4):46-52.
Research on small.scale, multi.angle face detection methods in complex backgrounds
Huang Jie,Liu Fen
School of Electronic Engineering, Tianjin University of Technology and Education
Abstract: A face detection approach which is named GhostNet.MTCNN was proposed to enhance the precision of small sized face detection in complex backgrounds. On the backbone of MTCNN, this approach uses the lower computational Ghost bottleneck module which is in the GhostNet to replace the convolutional function, and discards the common convolution which occupies computer resources to configure the network′s feature extraction function. Through the process, a new module will be set up. The experimental results showed that the approach can effectively balance parameter quantity and precision. Across three validation sets categorized as Easy, Medium and Hard, compared to the original MTCNN, the proposed GhostNet-MTCNN achieves notable improvements in accuracy respectively 5.6%, 6.6% and 7.8%, while the parameter quantity only with a minimal increase of 0.62M. Furthermore, compared to MobileNetV3-MTCNN, GhostNet-MTCNN outperforms by enhancing accuracy by 1.6%, 0.8% and 0.5%, meanwhile a reduction in parameter quantity by 1.27M. The study can not only enhance the precision of the module to detect the small-sized and multi-angle faces in complex backgrounds but also can effectively balance parameter quantity and detection precision, which will make it a superior choice for edge deployment devices.
Key words : face detection; multi-task cascaded convolutional networks; lightweight network; edge devices

引言

人臉檢測技術廣泛應用于考勤、解鎖設備、身份驗證、監(jiān)控場所、自動駕駛等場合[1-3]。在當前的人臉檢測領域,通常采用深度神經(jīng)網(wǎng)絡架構。2014年Girshick等人提出的R-CNN[4]目標檢測算法模型成功地將深度學習應用到目標檢測領域,這種目標檢測算法使用的是基于候選區(qū)域的檢測方法。Ren等人在FastR-CNN基礎上進行改進,提出了FasterR-CNN[5],該模型提出了專門的候選區(qū)域生成網(wǎng)絡。除了以上兩種目標檢測網(wǎng)絡模型外,還有基于單次目標檢測的網(wǎng)絡模型,如YOLO[6-8]和SSD[9]。這類方法優(yōu)勢在于檢測速度快,但對小目標的檢測效果不佳。這些深度神經(jīng)網(wǎng)絡在邊緣設備部署十分消耗資源,對于硬件的計算能力和能耗的要求很高,很難應用到實際場景中。多任務級聯(lián)卷積神經(jīng)網(wǎng)絡(Multi-task Cascaded Convolutional Networks,MTCNN)[10]作為一種經(jīng)典的人臉檢測方法,以其高效的性能、模型復雜度低而聞名,更適合邊緣設備的應用。但隨著人臉檢測任務的不斷復雜化,MTCNN也面臨一系列挑戰(zhàn),例如在小尺寸、遮擋、多角度和光照變化等情況下的檢測效果下降。文獻[11]中將MTCNN與VGGNet相結合,提升了模型檢測精度,但是相對應的模型計算量也變多了。


本文詳細內(nèi)容請下載:

http://ihrv.cn/resource/share/2000005968


作者信息:

黃杰,劉芬

(天津職業(yè)技術師范大學電子工程學院 ,天津300222)


Magazine.Subscription.jpg

此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權禁止轉載。