【正文】
2 Salient local image regions detection Researches on biological vision system indicate that anism (like drosophila) often pays attention to certain special regions in the scene for their behavioral relevance or local image cues while observing surroundings [8]. These regions can be taken as natural landmarks to effectively represent and distinguish different environments. Inspired by those, we use centersurround difference method to detect salient regions in multiscale image spaces. The opponencies of color and texture are puted to create the saliency map. Followup, subimage centered at the salient position in S is taken as the landmark region. The size of the landmark region can be decided adaptively according to the changes of gradient orientation of the local image [11]. Mobile robot navigation requires that natural landmarks should be detected stably when environments change to some extent. To validate the repeatability on landmark detection of our approach, we have done some experiments on the cases of scale, 2D rotation and viewpoint changes etc. shows that the door is detected for its saliency when viewpoint changes. More detailed analysis and results about scale and rotation can be found in our previous works[12]. 3 Scene recognition and localization Different from other scene recognition systems, our system doesn?t need training offline. In other words, our scenes are not classified in advance. When robot wanders, scenes captured at intervals of fixed time are used to build the vertex of a topological map, which represents the place where robot locates. Although the map?s geometric layout is ignored by the localization system, it is useful for visualization and debugging[13] and beneficial to path planning. So localization means searching the best match of current scene on the map. In this paper hidden Markov model is used to anize the extracted landmarks from current scene and create the vertex of topological map for its partial information resuming ability. Resembled by panoramic vision system, robot looks around to get omniimages. From Experiment on viewpoint changes each image, salient local regions are detected and formed to be a sequence, named as landmark sequence whose order is the same as the image sequence. Then a hidden Markov model is created based on the landmark sequence involving k salient local image regions, which is taken as the description of the place where the robot locates. In our system EVID70 camera has a view field of 177。 該系統(tǒng)使用單眼相機(jī)獲取機(jī)器人所處位置的全方位的 礦井環(huán)境圖像。這些 技術(shù) 貢獻(xiàn)使系統(tǒng) 具有 處理 比率變化、 二維旋轉(zhuǎn)和 視角變化的能力 。 定位識別 是這個(gè)領(lǐng)域的基本問題。 在訓(xùn)練階段,機(jī)器人收集 其所工作 環(huán)境 的圖像 , 并處理這些圖像 提取 出能表征該場景的 全局特征。 例如,周等人用多維直方圖來描述全 局 外觀特征。 這些特征對于 圖像縮放 、 平移 、 旋轉(zhuǎn)和 局部 光照 不 變 是穩(wěn)定的 。在實(shí)驗(yàn)中, NN在 表達(dá)兩種 部分 之間的相似性 時(shí)效果并不足夠好 。興趣點(diǎn)數(shù)量有效 減少 ,這使得處理更加容易。 2 局部圖像區(qū)域不變形的檢測 生物視覺系統(tǒng)的研究表明,生物體(像果蠅)在觀察周圍環(huán)境時(shí),經(jīng)常因?yàn)樗麄兊男袨榱?xí)慣注意場景中確定的特殊區(qū)域或者局部圖像信息。 隨后,以地圖突出位置為中心的分圖像,被定義為位置標(biāo)志區(qū)域。圖 1表明當(dāng)視角變化時(shí)因?yàn)樗耐怀鲂Ч箝T能被檢測出來。 當(dāng)機(jī)器人 徘徊時(shí) ,在固定時(shí)間間隔 內(nèi) 捕獲 的場景 用于生成拓?fù)?地圖的頂點(diǎn) ,它 表示了 機(jī)器人 所在位置 。 類似于全景視覺系統(tǒng),機(jī)器人環(huán)顧四周,以獲得全方位的圖像。 170 176。 讓 這 8個(gè)圖像作為隱藏 狀態(tài) Si( 1≤ i≤8 ),創(chuàng)建的 HMM模型可以由圖 3描述出來 。 距離 可以 根據(jù) 里程計(jì)的讀數(shù) 得到 。 已經(jīng) 證明光照和旋轉(zhuǎn)的變化可能 對 它 有更少的 影響。通常近鄰 方法( 神經(jīng)網(wǎng)絡(luò) )是用來測量兩 部分 之間的相似性。 匹配結(jié)果如 圖 5所示 。 圖像分辨率設(shè)置為 400 320 , 采樣頻率設(shè)置為 10幀 /秒 。 從 每個(gè)環(huán)境 選取 10個(gè)場景 ,并 為每個(gè)場景創(chuàng)建隱馬爾可夫模型。 HMM模型的參數(shù) 重復(fù)學(xué)習(xí)。 2)模糊邏輯用來識別局部圖像,并且強(qiáng)調(diào)個(gè)體特征對識別的作用,從而提高了位置標(biāo)志的可靠性 3) HMM模型是用來捕獲 圖像 結(jié)構(gòu)或那些 局部圖像之間的聯(lián)系 ,這 使場景識別問題 轉(zhuǎn)化為對 HMM的評價(jià)