JP4168793B2 - Layout evaluation system, layout evaluation program, and layout evaluation method - Google Patents

Description

【００７０】
図２は、誘導場の強さを求める際の遮蔽条件を説明する図である。
上式（１）を用いることにより、任意のディジタル画像の誘導場を求めることができる。また、曲線が複数ある場合、点Ｐにおける誘導場の強さＭ_xyは、個々の曲線が点Ｐにつくる誘導場の和になる。なお、上式（１）は、点Ｐから発した光が直接当たる部分のみ和をとるという制約条件がつく。例えば、点Ｐに対して、曲線ｆ₁(ｓ)，ｆ₂(ｓ)，ｆ₃(ｓ)が図２に示すように存在しているとすると、点Ｐから見えない部分、つまり、この場合、曲線ｆ₁(ｓ)に遮蔽されて点Ｐから見えない範囲Ｚに存在する部分の和はとらない。図２の例では、曲線ｆ₃(ｓ)のすべてと曲線ｆ₂(ｓ)の一部の和はとらないことになる。これを遮蔽条件という。
【００７１】
図３は、文字「Ａ」の誘導場の例であり、同図（ａ）は遮蔽条件を考慮して誘導場を求めた場合、同図（ｂ）は遮蔽条件を考慮しないで誘導場を求めた場合を示す図である。
図３（ａ）は、「Ａ」という文字について、上式（１）で計算した誘導場の例を示すものである。図３（ａ）の文字「Ａ」周辺に地図の等高線状に分布している細い線Ｌが誘導場の等ポテンシャル線であり、中央から外に行くほど誘導場の強さＭ_xyは弱くなりやがて０に近づく。
【００７２】
図３（ａ）の誘導場の分布の形状・強さにおける特徴、特に「Ａ」の頂点付近の分布が他より鋭角な特徴は、参考論文による四角形や三角形など、図形の角付近に関する誘導場の分布の心理実験結果と一致する。
また、図３（ｂ）は、遮蔽条件がなく、画素すべてを正電荷１の点電荷と仮定した誘導場の例であるが、誘導場の分布は、全体的に丸くなり、参考論文による心理実験結果と異なったものとなる。このように、遮蔽条件は、誘導場を特徴づける上で重要なものとなる。
【００７３】
このようにして、ある文字についての誘導場を得ることができる。なお、誘導場を用いた技術の例としては、例えば、「長石道博：「視覚の誘導場を用いた読みやすい和文プロポーショナル表示」、映像メディア学会誌、Vol.52，No.12，pp.1865-1872(1998)」（以下、第１の論文という。）や、「三好正純、下塩義文、古賀広昭、井手口健：「視覚の誘導場理論を用いた感性にもとづく文字配置の設計」、電子情報通信学会論文誌、82-A，9，1465-1473（1999)」（以下、第２の論文という。）がある。
【００７４】
本実施の形態では、このような誘導場を利用して、文字や写真、絵、図形などからなるひとまとまりの画像を、ある限られた所定の表示範囲内に表示する際、そのレイアウトが最適なものとなるようにレイアウトの評価を行う。
本実施の形態では、レイアウトの良し悪しを評価する際、レイアウト対象となるひとまとまりの画像を１つの誘導場計算対象とみなして、その誘導場を計算し、それによって求められた等ポテンシャル線の形状に基づいてレイアウトの良し悪しを評価する。
【００７５】
図４は、基準となるレイアウト例としてのある新聞記事の一部分の画像を示す図である。
いま、レイアウト対象となるひとまとまりの画像が図４に示されるように、文字列と写真からなる画像であるとする。図４に示される画像は、新聞記事の一部を示すもので、文字列部分Ｃと写真Ｐ１，Ｐ２からなり、図４に示されるレイアウトは、新聞紙面専門のデザイナによってなされたものであり、多くの人が見やすく内容の理解がしやすいとされるレイアウトであるとする。
【００７６】
図５は、図４に示す画像に対し、文字列部分はそれぞれの文字列を単純な線で表し、写真は単に矩形枠で表して誘導場を計算し、計算された誘導場から得られた等ポテンシャル線を示す図である。
図４に示すように、ある限られた表示範囲にレイアウトされるひとまとまりの画像全体について、上式（１）を用いて誘導場を計算すると、求められた誘導場によって、図５のような等ポテンシャル線Ｌが描かれる。なお、このようなレイアウト対象となる情報全体について誘導場を計算する際、図４で示した文字列部分Ｃは、図５に示すように、それぞれの文字列を単純な線で表し、写真Ｐ１，Ｐ２は、その外形を矩形枠で表して誘導場を計算する。
【００７７】
これは、レイアウトが各要素の位置関係や大きさで決まるため、各要素を単純化して表現することができるからであり、このように、各要素を単純化して表現した状態で誘導場を計算し、求められた誘導場から等ポテンシャル線を描けば、その等ポテンシャル線は、そのレイアウト全体の等ポテンシャル線を表すことができる。
【００７８】
なお、図４に示すレイアウトは、専門のデザイナによってデザインされた見やすく内容の理解がしやすいとされるレイアウトであり、このようにレイアウトされた画像全体から得られた等ポテンシャル線Ｌは、全体に凹凸が少なく丸みを帯びたものとなる。
このことから、レイアウト対象となるひとまとまりの画像全体について誘導場を計算し、それによって得られた等ポテンシャル線の形状から、その画像のレイアウトの良し悪しを判定することができる。つまり、得られた等ポテンシャル線の凹凸の度合いがわかれば、それによって当該画像のレイアウトが良いレイアウトであるかどうかの評価を行うことができる。
【００７９】
そこで、本実施の形態では、等ポテンシャル線の凹凸の度合いを等ポテンシャル線の複雑度として求め、その複雑度を当該画像のレイアウトの良し悪しを評価する指標として用いる。つまり、等ポテンシャル線が、凹凸が少なく丸みを帯びていればいるほど複雑度は小さくなり、等ポテンシャル線の凹凸が激しいほど複雑度は大きくなる。複雑度は、ｉ番目の等ポテンシャル線の複雑度をＣ_iで表せば、下式（２）により定義することができる。上式（２）において、Ｌ_iはｉ番目の等ポテンシャル線の長さ、Ｓ_iはｉ番目の等ポテンシャル線で囲まれた面の面積を表している。なお、ｉ番目の等ポテンシャル線の長さＬ_iは、そのポテンシャル線を構成するドット数と考えることができ、ｉ番目の等ポテンシャル線で囲まれた面の面積Ｓ_iは、ｉ番目の等ポテンシャル線で囲まれた面に存在するドット数と考えることができる。
【００８０】
【数２】

【００８１】
上式（２）によれば、レイアウト対象となるひとまとまりの画像について計算された誘導場によって描かれた等ポテンシャル線の長さが長いほど（凹凸が激しいほど）複雑度Ｃ_iの値は大きくなるといえる。逆に言えば、等ポテンシャル線に凹凸が少なく円に近いほど複雑度Ｃ_iは小さな値となる。
図６は、図４で示した基準レイアウトとその基準レイアウトを種々変化させたレイアウトとした場合の図である。
【００８２】
ここで、図４で示したひとまとまりの画像を図６で示すように色々なレイアウトとしたときのそれぞれの複雑度を計算してみる。図６では、図５と同様に、文字列部分Ｃはそれぞれの文字列を単純な線で表し、写真Ｐ１，Ｐ２は単に矩形枠で表している。
図６において、同図（ａ）は、図４と同じレイアウト（これをレイアウトＡ１という。）であり、同図（ｂ）は、図４の写真Ｐ２を文字列の中に配置したレイアウト（これをレイアウトＡ２という。）、同図（ｃ）は、写真Ｐ１が右下、写真Ｐ２が左上となっているレイアウト（これをレイアウトＡ３という。）、同図（ｄ）は、２つの写真Ｐ１，Ｐ２を文字列の中に配置したレイアウト（これをレイアウトＡ４という。）である。
【００８３】
図７は、図６（ａ）〜（ｄ）のようなレイアウトとしたときのそれぞれのレイアウトに対する複雑度を示す図である。
これらについて、まず、それぞれの誘導場を計算し、求められた誘導場によって描かれた等ポテンシャル線（それぞれのｉ番目のポテンシャル線）から、上式（２）によってそれぞれ複雑度を計算すると、図７のような結果が得られた。図７は、横軸にそれぞれのレイアウトＡ１〜Ａ６をとり、縦軸にそれぞれのレイアウトＡ１〜Ａ６に対して求められた複雑度をとっている。
【００８４】
図７によれば、デザイナによってレイアウトされた読みやすく内容の理解のしやすいとされるレイアウトＡ１（基準レイアウトＡ１という。）の複雑度が最も小さく、他の３つのレイアウトＡ２，Ａ３，Ａ４はいずれも、基準レイアウトＡ１に比べると、その複雑度は大きな値となっている。特に、この例においては、レイアウトＡ３が最も大きな複雑度となっている。
【００８５】
これは、前述したように、基準レイアウトＡ１から求められた誘導場に凹凸が少なく全体的に丸みを帯びているためであり、他の３つのレイアウトＡ２〜Ａ４はそれぞれのレイアウトから求められた等ポテンシャル線に凹凸が大きいためである。
また、等ポテンシャル線を利用し、画像全体における誘導場のエネルギＥは、下式（３）により定義することができる。下式（３）において、ｉはｉ番目の等ポテンシャル線を、Ｓ_iはｉ番目の等ポテンシャル線で囲まれた面の面積を、Ｐ_iはｉ番目の等ポテンシャル線におけるポテンシャル値をそれぞれ表している。これは、誘導場を３次元的に考えたとき、その誘導場の体積を求めるのに相当し、その体積の大きさをエネルギと定義している。
【００８６】
【数３】

【００８７】
以上は、新聞などの記事（多くは文字列と写真などからなる）の一部をレイアウト対象のひとまとまりの画像とし、そのひとまとまりの画像をレイアウトする場合についての評価を行った場合であるが、レイアウト対象の画像としては、一般的な画像を用いた場合の評価も同様に考えることができる。
次に、本発明に係るレイアウト評価装置の構成を図８を参照しながら説明する。
【００８８】
図８は、本発明に係るレイアウト評価装置の構成を示す機能ブロック図である。
本発明に係るレイアウト評価装置は、図８に示すように、レイアウト結果を評価する評価部１００と、読み手の注目箇所を考慮した見やすいレイアウト結果を学習する学習部２００とで構成されている。具体的には、ＣＰＵ、ＲＯＭ、ＲＡＭおよびＩ／Ｆ等をバス接続した一般的なコンピュータとして構成し、ＣＰＵは、ＲＯＭの所定領域に格納されている所定のプログラムを起動させ、そのプログラムに従って評価部１００および学習部２００として実現される処理を実行する。
【００８９】
評価部１００は、レイアウト結果を記憶したレイアウト結果記憶部１１０と、レイアウト結果記憶部１１０のレイアウト結果に基づいてレイアウト結果の注目領域を抽出する注目領域抽出部１１５と、注目領域抽出部１１５で抽出した注目領域について誘導場に関する特徴量を算出する特徴量算出部１２０と、注目領域の分布とレイアウト結果の評価との関係を注目領域分布モデルとして記憶した注目領域分布モデル記憶部１３０と、特徴量算出部１２０で算出した特徴量および注目領域分布モデル記憶部１３０の注目領域分布モデルに基づいてレイアウト結果記憶部１１０のレイアウト結果を評価するレイアウト結果評価部１４０とで構成されている。
【００９０】
注目領域抽出部１１５は、レイアウト結果記憶部１１０から評価対象となるレイアウト結果を読み出し、読み出したレイアウト結果に基づいて誘目度を算出し、算出した誘目度に基づいて注目領域を抽出するようになっている。注目領域の抽出は、上記方法により行う。複数の注目領域を抽出したときは、誘目度の高い上位Ｎ個（Ｎは自然数）の注目領域を抽出する。以下、レイアウト結果を構成する各画素について算出した誘目度をＦ（ｘ，ｙ）とする。ｘ，ｙは、レイアウト結果における画素のＸ座標およびＹ座標を示す。
【００９１】
特徴量算出部１２０は、注目領域抽出部１１５で抽出した注目領域について、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥを特徴量として算出するようになっている。誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥは、レイアウト結果を白黒２値化処理した画像に基づいて算出する。本実施の形態では、誘目度Ｆ（ｘ、ｙ）並びに各特徴量Ｍ_xy，Ｃ_iおよびＥをそれぞれベクトルとして取り扱う。
【００９２】
図９は、ニューラルネットワーク３００の構成を示す図である。
注目領域分布モデル記憶部１３０は、図９に示すように、ニューラルネットワーク３００により注目領域の分布を記憶するようになっている。
ニューラルネットワーク３００は、図９に示すように、誘目度Ｆ（ｘ、ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥを入力するｉ個の入力層Ｉ_iと、各入力層Ｉ_iからの出力を入力するｊ個の中間層Ｈ_jと、各中間層Ｈ_jの出力を入力して評価値を出力する出力層Ｏ_kとから構成されている。そして、入力層Ｉ_iと中間層Ｈ_jとは結合係数Ｗ_ijのシナプスにより、中間層Ｈ_jと出力層Ｏ_kとは結合係数Ｗ_jkのシナプスによりそれぞれ結合されている。
【００９３】
また、ニューラルネットワーク３００は、後述の特徴学習部２３０により、読み手の注目箇所を考慮した見やすいレイアウト結果の特徴を学習している。したがって、読み手の注目箇所を考慮した見やすいレイアウト結果から算出した誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥをニューラルネットワーク３００に入力したときは、評価値として比較的高い値が出力層Ｏ_kから出力され、読み手の注目箇所を考慮しないレイアウト結果から算出した誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥをニューラルネットワーク３００に入力したときは、評価値として比較的低い値が出力層Ｏ_kから出力される。
【００９４】
図８に戻り、レイアウト結果評価部１４０は、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ，ｙ）、並びに特徴量算出部１２０で算出した特徴量Ｍ_xy，Ｃ_iおよびＥをニューラルネットワーク３００に入力し、ニューラルネットワーク３００の出力値をレイアウト結果の評価として出力するようになっている。出力した評価値は、ディスプレイ等の表示装置に表示してもよいし、レイアウト結果の選択やレイアウトの決定に利用することもできる。
【００９５】
学習部２００は、図８に示すように、レイアウト結果記憶部１１０のなかから学習対象とするレイアウト結果の指定を入力するレイアウト結果指定入力部２１０と、レイアウト結果記憶部１１０のレイアウト結果のうちレイアウト結果指定入力部２１０で入力した指定に係るものに基づいてレイアウト結果の注目領域を抽出する注目領域抽出部２１５と、注目領域抽出部２１５で抽出した注目領域について誘導場に関する特徴量を算出する特徴量算出部２２０と、特徴量算出部２２０で算出した特徴量に基づいて読み手の注目箇所を考慮した見やすいレイアウト結果の特徴を学習する特徴学習部２３０とで構成されている。
【００９６】
注目領域抽出部２１５は、注目領域抽出部１１５と同一機能を有して構成されており、レイアウト結果記憶部１１０のレイアウト結果のうちレイアウト結果指定入力部２１０で入力した指定に係るものに基づいて誘目度Ｆ（ｘ，ｙ）を算出し、算出した誘目度Ｆ（ｘ，ｙ）に基づいて注目領域を抽出するようになっている。
【００９７】
特徴量算出部２２０は、特徴量算出部１２０と同一機能を有して構成されており、注目領域抽出部１１５で抽出した注目領域について、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥを特徴量として算出するようになっている。
特徴学習部２３０は、注目領域抽出部２１５で算出した誘目度Ｆ（ｘ，ｙ）、並びに特徴量算出部２２０で算出した特徴量Ｍ_xy，Ｃ_iおよびＥに基づいて、公知のバックプロパゲーション法その他の学習法によりニューラルネットワーク３００を学習するようになっている。学習では、レイアウト結果指定入力部２１０で入力した指定に係るレイアウト結果から算出した誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥをニューラルネットワーク３００に入力したときに、評価値として比較的高い値が出力層Ｏ_kから出力されるように結合係数Ｗ_ij，Ｗ_jkを決定する。例えば、バックプロパゲーション法を用いる場合は、前向き演算または後ろ向き演算により結合係数Ｗ_ij，Ｗ_jkを決定する。
【００９８】
次に、本実施の形態の動作を説明する。
まず、ニューラルネットワーク３００を学習する場合を説明する。
ニューラルネットワーク３００を学習する場合、ユーザまたはデザイナは、レイアウト結果記憶部１１０のなかから読み手の注目箇所を考慮した見やすいレイアウト結果をいくつか指定する。この指定は、レイアウト結果指定入力部２１０に入力する。
【００９９】
レイアウト結果の指定が入力されると、注目領域抽出部２１５により、レイアウト結果記憶部１１０のレイアウト結果のうち入力された指定に係るものに基づいて誘目度Ｆ（ｘ，ｙ）が算出され、算出された誘目度Ｆ（ｘ，ｙ）に基づいて注目領域が抽出される。次いで、特徴量算出部２２０により、抽出された注目領域について、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥが特徴量として算出される。そして、特徴学習部２３０により、算出された誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥに基づいてニューラルネットワーク３００が学習される。この一連の処理は、指定されたすべてのレイアウト結果について行われる。
【０１００】
次に、レイアウトを評価する場合を説明する。
図１０は、レイアウト対象となる画像６００，６０２を示す図である。
図１１は、評価対象となるレイアウト結果６１０，６２０を示す図である。
いま、図１０に示すように、レイアウト対象となる２つの画像６００，６０２がある。これら２つの画像を用いて異なるレイアウトパターンで２つのレイアウトを作成した。そのレイアウト結果がそれぞれ図１０（ａ），（ｂ）に示すようになる。レイアウト結果６１０は、図１０（ａ）に示すように、左上領域に画像６００を、右下領域に画像６０２をそれぞれ配置し、画像６００，６０２の大きさをほぼ同一としたものである。右上領域および左下領域には、画像６００，６０２に対応する文章が配置されている。また、レイアウト結果６２０は、図１０（ｂ）に示すように、左上領域に画像６００を、右下領域に画像６０２をそれぞれ配置し、画像６００を画像６０２よりも大きくし主体的に配置したものである。右上領域および左下領域には、画像６００，６０２に対応する文章が配置されている。
【０１０１】
図１２は、レイアウト結果６１０，６２０の注目領域を示す図である。
まず、注目領域抽出部１１５により、レイアウト結果記憶部１１０からレイアウト結果６１０が読み出され、読み出されたレイアウト結果６１０に基づいて誘目度Ｆ（ｘ，ｙ）が算出され、算出された誘目度Ｆ（ｘ，ｙ）に基づいて注目領域が抽出される。複数の注目領域が抽出されたときは、誘目度Ｆ（ｘ，ｙ）が高い上位Ｎ個（例えば４つ）の注目領域が抽出される。レイアウト結果６１０の注目領域は、図１２（ａ）に示すように、左上領域、右上領域、左下領域および右下領域においてそれぞれ斜線で示す領域として抽出される。なお、説明の便宜上、注目領域を楕円で示しているが、実際には、領域分割の手法によって領域の形状は変化する。
【０１０２】
次いで、特徴量算出部１２０により、レイアウト結果６１０の各注目領域について、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥが特徴量として算出される。そして、レイアウト結果評価部１４０により、算出された誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥがニューラルネットワーク３００に入力され、ニューラルネットワーク３００の出力値がレイアウト結果の評価として出力される。
【０１０３】
同様に、注目領域抽出部１１５により、レイアウト結果記憶部１１０からレイアウト結果６２０が読み出され、読み出されたレイアウト結果６２０に基づいて誘目度Ｆ（ｘ，ｙ）が算出され、算出された誘目度Ｆ（ｘ，ｙ）に基づいて注目領域が抽出される。複数の注目領域が抽出されたときは、誘目度Ｆ（ｘ，ｙ）が高い上位Ｎ個の注目領域が抽出される。レイアウト結果６２０の注目領域は、図１２（ｂ）に示すように、左上領域、右上領域、左下領域および右下領域においてそれぞれ斜線で示す領域として抽出される。レイアウト結果６１０と異なるのは、画像６００に対応する注目領域が大きく、画像６０２に対応する注目領域が小さい点である。
【０１０４】
次いで、特徴量算出部１２０により、レイアウト結果６２０の各注目領域について、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥが特徴量として算出される。そして、レイアウト結果評価部１４０により、算出された誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥがニューラルネットワーク３００に入力され、ニューラルネットワーク３００の出力値がレイアウト結果の評価として出力される。
【０１０５】
このようにして、本実施の形態では、レイアウト結果に基づいてレイアウト結果の注目領域を抽出する注目領域抽出部１１５と、注目領域抽出部１１５で抽出した注目領域の分布に基づいてレイアウト結果を評価するレイアウト結果評価部１４０とを備える。
これにより、注目領域に基づいてレイアウト結果が評価されるので、評価結果をレイアウトに反映させれば、従来に比して、読み手の注目箇所を考慮した見やすいレイアウトを実現することができる。
【０１０６】
さらに、本実施の形態では、注目領域抽出部１１５で抽出した注目領域について誘導場に関する特徴量を算出する特徴量算出部１２０を備え、レイアウト結果評価部１４０は、特徴量算出部１２０で算出した特徴量に基づいてレイアウト結果を評価するようになっている。
これにより、生理学、心理学的な知見に基づく誘導場をレイアウトの評価に利用したことにより、評価結果をレイアウトに反映させれば、比較的見栄えのよいレイアウトを実現することができる。
【０１０７】
さらに、本実施の形態では、誘目度Ｆ（ｘ、ｙ）とレイアウト結果の評価との関係をニューラルネットワーク３００に学習させておき、レイアウト結果評価部１４０は、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ、ｙ）をニューラルネットワーク３００に入力し、ニューラルネットワーク３００の出力値をレイアウト結果の評価として出力するようになっている。
【０１０８】
これにより、読み手の注目箇所を考慮した見やすいレイアウト結果を用意し、そのレイアウト結果について誘目度Ｆ（ｘ、ｙ）の分布をニューラルネットワーク３００に学習させておけば、レイアウト結果の良し悪しを比較的容易に判定することができる。
さらに、本実施の形態では、誘導場の強さＭ_xyを示す特徴量とレイアウト結果の評価との関係をニューラルネットワーク３００に学習させておき、特徴量算出部１２０は、注目領域抽出部１１５で抽出した注目領域について誘導場の強さＭ_xyを算出し、レイアウト結果評価部１４０は、特徴量算出部１２０で算出した誘導場の強さＭ_xyをニューラルネットワーク３００に入力し、ニューラルネットワーク３００の出力値をレイアウト結果の評価として出力するようになっている。
【０１０９】
これにより、生理学、心理学的な知見に基づく誘導場の強さＭ_xyをレイアウトの評価に利用したことにより、評価結果をレイアウトに反映させれば、さらに見栄えのよいレイアウトを実現することができる。また、読み手の注目箇所を考慮した見やすいレイアウト結果を用意し、そのレイアウト結果について各注目領域ごとに誘導場の強さＭ_xyを示す特徴量をニューラルネットワーク３００に学習させておけば、レイアウト結果の良し悪しを比較的容易に判定することができる。
【０１１０】
さらに、本実施の形態では、等ポテンシャル線の複雑度Ｃ_iを示す特徴量とレイアウト結果の評価との関係をニューラルネットワーク３００に学習させておき、特徴量算出部１２０は、注目領域抽出部１１５で抽出した注目領域について誘導場を算出し、算出した誘導場から等ポテンシャル線を得て、その等ポテンシャル線の複雑度Ｃ_iを算出し、レイアウト結果評価部１４０は、特徴量算出部１２０で算出した複雑度Ｃ_iをニューラルネットワーク３００に入力し、ニューラルネットワーク３００の出力値をレイアウト結果の評価として出力するようになっている。
【０１１１】
これにより、生理学、心理学的な知見に基づく誘導場における等ポテンシャル線の複雑度Ｃ_iをレイアウトの評価に利用したことにより、評価結果をレイアウトに反映させれば、さらに見栄えのよいレイアウトを実現することができる。また、読み手の注目箇所を考慮した見やすいレイアウト結果を用意し、そのレイアウト結果について各注目領域ごとに等ポテンシャル線の複雑度Ｃ_iを示す特徴量をニューラルネットワーク３００に学習させておけば、レイアウト結果の良し悪しを比較的容易に判定することができる。
【０１１２】
さらに、本実施の形態では、誘導場のエネルギＥを示す特徴量とレイアウト結果の評価との関係をニューラルネットワーク３００に学習させておき、特徴量算出部１２０は、注目領域抽出部１１５で抽出した注目領域について誘導場のエネルギＥを算出し、レイアウト結果評価部１４０は、特徴量算出部１２０で算出した誘導場のエネルギＥをニューラルネットワーク３００に入力し、ニューラルネットワーク３００の出力値をレイアウト結果の評価として出力するようになっている。
【０１１３】
これにより、生理学、心理学的な知見に基づく誘導場のエネルギＥをレイアウトの評価に利用したことにより、評価結果をレイアウトに反映させれば、さらに見栄えのよいレイアウトを実現することができる。また、読み手の注目箇所を考慮した見やすいレイアウト結果を用意し、そのレイアウト結果について各注目領域ごとに誘導場のエネルギＥを示す特徴量をニューラルネットワーク３００に学習させておけば、レイアウト結果の良し悪しを比較的容易に判定することができる。
【０１１４】
さらに、本実施の形態では、レイアウト結果評価部１４０は、注目領域抽出部１１５で複数の注目領域を抽出したときは、複数の注目領域のうち上位Ｎ個のものの分布に基づいてレイアウト結果を評価するようになっている。
これにより、複数の注目領域のうち上位Ｎ個のものの分布に基づいてレイアウト結果が評価されるので、評価結果をレイアウトに反映させれば、読み手の注目箇所を考慮したさらに見やすいレイアウトを実現することができるとともに、注目領域が多数存在した場合に処理負荷を低減することができる。
【０１１５】
上記第１の実施の形態において、注目領域抽出部１１５は、発明３ないし１０、１９若しくは２１の注目領域抽出手段、または発明２の誘目度算出手段に対応し、注目領域抽出部１１５による抽出は、発明２３の注目領域抽出ステップに対応し、特徴量算出部１２０は、発明５ないし８、１１ないし１８の特徴量算出手段に対応している。また、注目領域分布モデル記憶部１３０は、発明９、１１、１３、１５または１７のレイアウト評価モデル記憶手段に対応し、レイアウト結果評価部１４０は、発明２ないし５、９ないし１９または２１のレイアウト結果評価手段に対応し、レイアウト結果評価部１４０による評価は、発明２３のレイアウト結果評価ステップに対応している。
【０１１６】
また、上記第１の実施の形態において、ニューラルネットワーク３００は、発明９ないし１８のレイアウト評価モデルに対応している。
次に、本発明の第２の実施の形態を図面を参照しながら説明する。図１３は、本発明に係るレイアウト評価システムおよびレイアウト評価プログラム、並びにレイアウト評価方法の第２の実施の形態を示す図である。
【０１１７】
本実施の形態は、本発明に係るレイアウト評価システムおよびレイアウト評価プログラム、並びにレイアウト評価方法を、読み手の注目箇所を考慮してレイアウト結果を評価する場合について適用したものであり、上記第１の実施の形態と異なるのは、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ，ｙ）に基づいてレイアウト結果を評価する点にある。なお、以下、上記第１の実施の形態と異なる部分についてのみ説明し、上記第１の実施の形態と重複する部分については同一の符号を付して説明を省略する。
【０１１８】
次に、本発明に係るレイアウト評価装置の構成を図１３を参照しながら説明する。
図１３は、本発明に係るレイアウト評価装置の構成を示す機能ブロック図である。
本発明に係るレイアウト評価装置は、図１３に示すように、レイアウト結果記憶部１１０と、注目領域抽出部１１５と、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ，ｙ）に基づいてレイアウト結果を評価するレイアウト結果評価部１５０とで構成されている。
【０１１９】
レイアウト結果評価部１５０は、下式（４）により、注目領域抽出部１１５で抽出した誘目度Ｆ（ｘ，ｙ）にガウス関数Ｇ（ｘ，ｙ）による重み付けを行い、レイアウト評価値Ｈを算出し、算出したレイアウト評価値Ｈをレイアウト結果の評価として出力するようになっている。
【０１２０】
【数４】

【０１２１】
次に、本実施の形態の動作を説明する。
まず、注目領域抽出部１１５により、レイアウト結果記憶部１１０からレイアウト結果６１０が読み出され、読み出されたレイアウト結果６１０に基づいて誘目度Ｆ（ｘ，ｙ）が算出され、算出された誘目度Ｆ（ｘ，ｙ）に基づいて注目領域が抽出される。そして、レイアウト結果評価部１５０により、上式（４）によりレイアウト評価値Ｈが算出され、算出されたレイアウト評価値Ｈがレイアウト結果の評価として出力される。
【０１２２】
このようにして、本実施の形態では、レイアウト結果に基づいて誘目度Ｆ（ｘ，ｙ）を算出する注目領域抽出部１１５と、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ，ｙ）に基づいてレイアウト結果を評価するレイアウト結果評価部１５０とを備える。
これにより、誘目度に基づいてレイアウト結果が評価されるので、評価結果をレイアウトに反映させれば、従来に比して、読み手の注目箇所を考慮した見やすいレイアウトを実現することができる。
【０１２３】
上記第２の実施の形態において、注目領域抽出部１１５は、発明３、４若しくは２１の注目領域抽出手段、または発明２の誘目度算出手段に対応し、注目領域抽出部１１５による抽出は、発明２３の注目領域抽出ステップに対応し、レイアウト結果評価部１５０は、発明２ないし４または２１のレイアウト結果評価手段に対応している。また、レイアウト結果評価部１５０による評価は、発明２３のレイアウト結果評価ステップに対応している。
【０１２４】
なお、上記第１の実施の形態において、レイアウト結果評価部１４０は、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ，ｙ）、並びに特徴量算出部１２０で算出した特徴量Ｍ_xy，Ｃ_iおよびＥをニューラルネットワーク３００に入力し、ニューラルネットワーク３００の出力値をレイアウト結果の評価として出力するように構成したが、これに限らず、レイアウト結果評価部１４０は、ニューラルネットワーク３００を用いず、注目領域抽出部１１５で算出した誘目度Ｆ（ｘ，ｙ）、並びに特徴量算出部１２０で算出した特徴量Ｍ_xy，Ｃ_iおよびＥに基づいてレイアウト結果を評価するように構成することもできる。例えば、それぞれの値の絶対値に対して所定の重み付けを行って加算することにより評価値を算出し、評価値の大小によりレイアウト結果を評価することができる。
【０１２５】
また、上記第１の実施の形態においては、ニューラルネットワーク３００の学習を行うにあたって、読み手の注目箇所を考慮した見やすいレイアウト結果をレイアウト結果記憶部１１０のなかから選択するように構成したが、これに限らず、一般的に見やすいレイアウト結果をレイアウト結果記憶部１１０のなかから選択するように構成してもよい。この場合、複数のユーザまたはデザイナに見やすいと思うレイアウト結果を指定してもらい、指定されたレイアウト結果の特徴を、上記第１の実施の形態と同じ要領でニューラルネットワーク３００に学習させておけばよい。
【０１２６】
さらに、この場合、複数のユーザまたはデザイナに印象の良し／悪しを入力してもらうだけでなく、印象の強い／弱いを入力してもらい、これに基づいてニューラルネットワーク３００に学習させることも可能である。これにより、一般的な評価が学習できるため、複数の人の好みに適合したレイアウト評価を行うことができる。
【０１２７】
さらに、この場合、例えば、１０代、２０代、３０代など、年齢に応じてグループ分けして、各グループごとに、そのユーザまたはデザイナに印象の良いと思うレイアウト結果を指定してもらい、指定されたレイアウト結果の特徴をニューラルネットワーク３００に学習させることも可能である。これにより、同世代の人の好みに適合したレイアウト評価を行うことができる。また、あるレイアウト結果が何代の人に好まれるかを調べることにも使用できる。
【０１２８】
また、上記第１の実施の形態において、ニューラルネットワーク３００は、出力層Ｏ_kを１つだけ設けて構成したが、これに限らず、複数の出力層を設けて構成してもよい。例えば、ユーザの好き／嫌いのいずれかを出力する第１の出力層と、ユーザの印象の良し／悪しのいずれかを出力する第２の出力層と、ユーザの印象の強さ／弱さのいずれかを出力する第３の出力層とを設けて構成することもできる。
【０１２９】
また、上記第１の実施の形態において、特徴量算出部１２０，２２０は、注目領域を構成するすべての画素の特徴量を算出し、算出した特徴量に基づいて学習を行うように構成したが、これに限らず、注目領域を構成する画素のうち、例えば、縦方向５つ横方向５つの画素からなる矩形領域の画素群において４つ角の画素を対象とし、対象画素の特徴量（例えば、平均値）を算出し、算出した特徴量に基づいて学習を行うように構成してもよい。
【０１３０】
また、上記第１の実施の形態においては、注目領域について算出した誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥに基づいてレイアウト評価および学習を行うように構成したが、これに限らず、注目領域について算出した誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥのうちいずれかに基づいてレイアウト評価および学習を行うように構成することもできる。
【０１３１】
また、上記第１の実施の形態において、レイアウト結果評価部１４０は、注目領域について算出した誘目度Ｆ（ｘ，ｙ）並びに特徴量Ｍ_xy，Ｃ_iおよびＥに基づいてレイアウト結果を評価するように構成したが、これに限らず、注目領域の総面積に基づいてレイアウト結果を評価するように構成することもできる。
また、上記第１の実施の形態においては、ニューラルネットワーク３００の学習法としてバックプロパゲーション法を例示したが、これに限らず、自己組織化による教師なし学習法を利用することもできる。これにより、例えば、ユーザが作成した２５個のレイアウト結果の特徴を学習し、そのレイアウト結果の傾向に沿って学習することができ、そのユーザの好みを自動的に学習することができる。
【０１３２】
また、上記第１の実施の形態においては、レイアウト結果を白黒２値化処理した画像に基づいて、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥを算出するように構成したが、これに限らず、カラーのレイアウト結果そのものに基づいて、誘導場の強さＭ_xy、等ポテンシャル線の複雑度Ｃ_iおよび誘導場のエネルギＥを算出するように構成することもできる。
【０１３３】
また、上記第１の実施の形態においては、ニューラルネットワーク３００を利用したが、これに代えて、「Naive BayesやBayesian Network」のような確率モデルを利用することもできる。このような構成であっても、上記第１の実施の形態と同等の効果が得られる。
また、上記第２の実施の形態において、注目領域抽出部１１５は、重み付けを行うことなく注目領域の誘目度Ｆ（ｘ，ｙ）を算出するように構成したが、これに限らず、注目領域の中心から外側に向かうにつれて減少するような重み付けを行うことにより注目領域の誘目度Ｆ（ｘ，ｙ）を算出するように構成することもできる。
【０１３４】
また、上記第２の実施の形態において、レイアウト結果評価部１５０は、上式（４）により、注目領域抽出部１１５で抽出した誘目度Ｆ（ｘ，ｙ）にガウス関数Ｇ（ｘ，ｙ）による重み付けを行い、レイアウト評価値Ｈを算出し、算出したレイアウト評価値Ｈをレイアウト結果の評価として出力するように構成したが、ガウス関数Ｇ（ｘ，ｙ）に代えて線形補間やシグモイド関数を用いることができる。
【０１３５】
また、上記第１の実施の形態においては、本発明に係るレイアウト評価システムおよびレイアウト評価プログラム、並びにレイアウト評価方法を、読み手の注目箇所を考慮した見やすいレイアウト結果をニューラルネットワーク３００にあらかじめ学習させておき、ニューラルネットワーク３００を用いて、与えられたレイアウト結果を評価する場合について適用したが、これに限らず、本発明の主旨を逸脱しない範囲で他の場合にも適用可能である。
【図面の簡単な説明】
【図１】 ディジタル画像の画素配列を示す図である。
【図２】 誘導場の強さを求める際の遮蔽条件を説明する図である。
【図３】 文字「Ａ」の誘導場の例であり、同図（ａ）は遮蔽条件を考慮して誘導場を求めた場合、同図（ｂ）は遮蔽条件を考慮しないで誘導場を求めた場合を示す図である。
【図４】 基準となるレイアウト例としてのある新聞記事の一部分の画像を示す図である。
【図５】 図４に示す画像に対し、文字列部分はそれぞれの文字列を単純な線で表し、写真は単に矩形枠で表して誘導場を計算し、計算された誘導場から得られた等ポテンシャル線を示す図である。
【図６】 図４で示した基準レイアウトとその基準レイアウトを種々変化させたレイアウトとした場合の図である。
【図７】 図６（ａ）〜（ｄ）のようなレイアウトとしたときのそれぞれのレイアウトに対する複雑度を示す図である。
【図８】 本発明に係るレイアウト評価装置の構成を示す機能ブロック図である。
【図９】 ニューラルネットワーク３００の構成を示す図である。
【図１０】 レイアウト対象となる画像６００，６０２を示す図である。
【図１１】 評価対象となるレイアウト結果６１０，６２０を示す図である。
【図１２】 レイアウト結果６１０，６２０の注目領域を示す図である。
【図１３】 本発明に係るレイアウト評価装置の構成を示す機能ブロック図である。
【符号の説明】
１００…評価部，１１０…レイアウト結果記憶部，１１５…注目領域抽出部，１２０…特徴量算出部，１３０…注目領域分布モデル記憶部，１４０，１５０…レイアウト結果評価部，２００…学習部，２１０…レイアウト結果指定入力部，２１５…注目領域抽出部，２２０…特徴量算出部，２３０…特徴学習部，３００…ニューラルネットワーク，６００，６０２…画像，６１０，６２０…レイアウト結果[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a system and program for evaluating a layout result of laying out information such as documents and images, and a method thereof, and more particularly to a layout evaluation system suitable for realizing an easy-to-view layout in consideration of a reader's attention point. The present invention relates to a layout evaluation program and a layout evaluation method.
[0002]
[Prior art]
Conventionally, as a technique for laying out publication information such as documents and images, for example, a character arrangement determining device disclosed in Patent Document 1 (hereinafter referred to as a first conventional example) and Patent Document 2 are disclosed. Layout evaluation apparatus (hereinafter referred to as second conventional example), article placement apparatus disclosed in Patent Document 3 (hereinafter referred to as third conventional example), and article placement apparatus disclosed in Patent Document 4 (Hereinafter referred to as the fourth conventional example).
[0003]
The first conventional example is an initial series / line number calculation unit for determining the number and series of lines for arranging characters based on formwork information and character information, and the basic set for each character from the calculation result, etc. Characters are placed on each line in order according to the character spacing, and the character number determination unit that makes a tentative determination of the number of characters and the tentatively determined character arrangement status of the last line are inspected to determine the excess or deficiency between characters in the tentatively determined number of characters. The character number changing unit that adds / deletes characters in a predetermined line and rearranges them so that the final line condition is good, and the character spacing between each line with the determined number of characters is It is composed of a character arrangement determining unit that determines the character spacing so that the ratio is the same according to the reference character spacing, and an inference unit that corrects the number of lines and series according to the evaluation result of the evaluation unit. The evaluation unit performs evaluation by fuzzy reasoning with four inputs and one output. For example, when the space between characters is wide, when the space between characters is narrow, when the space between lines is wide, when the space between lines is narrow, and when there is a variation between characters. In any case, a bad evaluation result is output. As a result, an arbitrary character string can be automatically arranged in a form with an arbitrary shape without a blank space and with a good balance.
[0004]
The second conventional example evaluates a state in which a plurality of blocks in which posted information is described are laid out in a predetermined layout area. Layout information including a plurality of blocks is stored in the storage unit. In the empty division unit of the layout evaluation unit, the empty area in which no block is laid out is subdivided into rectangular areas. A two-dimensional variance of the subdivided rectangular vacant areas is obtained by the evaluation value calculation unit and used as the evaluation value of the layout. The evaluation unit compares the evaluation value with a predetermined threshold value, and if the evaluation value is higher than the threshold value, it is displayed on the display unit as an appropriate layout. If the evaluation value is lower than the threshold value, an instruction to redo the layout is given. Thereby, it is possible to evaluate a state in which a plurality of blocks in which posted information is described are laid out in a predetermined area.
[0005]
In the third conventional example, the layout result is evaluated and printing is performed with the best layout. Arrangement unit combines single columns or adjacent columns that conform to the arrangement of each article according to random article processing order for several types of column configurations that are divided into several rectangular areas with random vertical and horizontal lines on the page The composite column is searched, and several layout results in which a plurality of articles are arranged are generated. The layout result evaluation unit assigns an evaluation value to each layout result based on the article registration priority of each article, and the best layout result search unit selects a layout result having the highest evaluation value. The printing unit prints a document in which a plurality of articles are arranged on a sheet according to the best layout result. Thus, it is possible to automatically lay out a document in which a plurality of articles are arranged so that more important articles can be listed within a limited layout area.
[0006]
The fourth conventional example performs layout of a document in which a plurality of pieces of article information are arranged like a newspaper or a magazine. An arrangement source consisting of a column arrangement process order generation unit and an article arrangement process order generation unit, and a column arrangement process order and an article arrangement process order. Generate multiple pieces of information. Articles are placed based on each placement source information, and each layout result is evaluated by the layout result evaluation unit so that the placement source information corresponding to a higher evaluation value is selected with a greater probability. The new arrangement source information is synthesized by combining the arrangement source information, and the arrangement and evaluation are performed in the same manner. The layout result evaluation unit inputs the layout result from the assigning unit, sums up the placement priority of the posted article from the input layout result and article placement priority information, and adds it to the evaluation value. After the repeated layout result is generated, articles are arranged based on the layout result corresponding to the maximum evaluation value. As a result, the layout can be performed relatively well without human intervention.
[0007]
[Patent Document 1]
JP-A-6-131331
[Patent Document 2]
JP-A-8-255063
[Patent Document 3]
JP-A-9-69096
[Patent Document 4]
JP 9-237273 A
[0008]
[Problems to be solved by the invention]
When publishing posted information such as a document or an image, it is desirable to lay out the posted information in consideration of a portion that the reader pays attention to in the layout result from the viewpoint of realizing an easy-to-view layout. It is very easy for the reader to see if there are noticeable points in the layout results. In order to realize this, for example, it is necessary to determine whether the layout is good or bad by using how much the portion of the layout result noticed by the reader is conspicuous.
[0009]
In the first conventional example, bad evaluation is performed on the layout result when the space between characters or lines is wide, when the space between characters or lines is narrow, or when the variation between characters is large. However, since the layout evaluation is not performed in consideration of the reader's attention, there is a problem that it is insufficient to realize an easy-to-see layout.
[0010]
In the second conventional example, the empty area is subdivided into rectangular areas, a two-dimensional distribution of the subdivided rectangular empty areas is obtained, and the distribution is used as an evaluation value of the layout result. However, similarly to the first conventional example, the layout evaluation is not performed in consideration of the reader's attention point, so that there is a problem that it is insufficient for realizing an easy-to-see layout.
[0011]
In the third and fourth conventional examples, the layout is evaluated based on the article registration priority of the article. However, similarly to the first conventional example, the layout evaluation is not performed in consideration of the reader's attention point, so that there is a problem that it is insufficient for realizing an easy-to-see layout.
Therefore, the present invention has been made paying attention to such an unsolved problem of the conventional technology, and a layout evaluation system and layout suitable for realizing an easy-to-view layout in consideration of the reader's attention location An object is to provide an evaluation program and a layout evaluation method.
[0012]
[Means for Solving the Problems]
[Invention 1]
In order to achieve the above object, the layout evaluation system of the invention 1 comprises:
A system for evaluating layout,
The layout result is evaluated based on the degree of attraction obtained from the layout result.
[0013]
With such a configuration, when a layout result is given, the layout result is evaluated based on the degree of attraction obtained from the given layout result.
The degree of attraction is a parameter that matches the human subjectivity, and is used to extract a part of the layout result that the reader pays attention to. Therefore, the layout result is evaluated based on the degree of attraction. Therefore, if the evaluation result is reflected in the layout, it is possible to realize an easy-to-view layout that takes into account the reader's attention as compared to the conventional case. It is done.
[0014]
[Invention 2]
Furthermore, the layout evaluation system of the invention 2 is the layout evaluation system of the invention 1,
An attraction degree calculation unit that calculates an attraction degree based on the layout result, and a layout result evaluation unit that evaluates the layout result based on the attraction degree calculated by the attraction degree calculation unit.
With such a configuration, the attraction degree calculation means calculates the attraction degree based on the layout result, and the layout result evaluation means evaluates the layout result based on the calculated attraction degree.
[0015]
[Invention 3]
Furthermore, the layout evaluation system of the invention 3 is
A system for evaluating layout,
An attention area extracting unit that extracts an attention area of the layout result based on a layout result; and a layout result evaluation unit that evaluates the layout result based on an extraction result of the attention area extraction unit.
[0016]
In such a configuration, when a layout result is given, the attention area extracting unit extracts a attention area of the layout result based on the given layout result, and the layout result evaluating unit based on the extraction result. The layout result is evaluated.
As a result, the layout result is evaluated based on the attention area. Therefore, if the evaluation result is reflected in the layout, it is possible to realize an easy-to-view layout that takes into account the reader's attention position compared to the conventional case. can get.
[0017]
Here, the attention area of the layout result refers to a portion of the layout result that the reader is likely to pay attention to. Hereinafter, the same applies to the layout evaluation program of the invention 21 and the layout evaluation method of the invention 23.
Further, the attention area extracting unit only needs to extract the attention area of the layout result based on the layout result. For example, the degree of attraction can be calculated based on the layout result, and the attention area of the layout result can be extracted based on the calculated degree of attraction. The same applies to the layout evaluation program of the invention 21 below.
[0018]
[Invention 4]
Furthermore, the layout evaluation system of the invention 4 is the layout evaluation system of the invention 3,
The layout result evaluation means is characterized in that the layout result is evaluated based on the distribution of the attention area extracted by the attention area extraction means.
With such a configuration, the layout result is evaluated by the layout result evaluation means based on the extracted distribution of the attention area.
As a result, the layout result is evaluated based on the distribution of the attention area, and if the evaluation result is reflected in the layout, it is possible to realize an even easier-to-view layout in consideration of the reader's attention position.
[0019]
[Invention 5]
Furthermore, the layout evaluation system of the invention 5 is the layout evaluation system of any of the inventions 3 and 4,
And a feature amount calculating means for calculating a feature amount relating to a visual guidance field for the attention area extracted by the attention area extracting means,
The layout result evaluation unit is characterized in that the layout result is evaluated based on the feature amount calculated by the feature amount calculation unit.
[0020]
With such a configuration, the feature amount calculation unit calculates the feature amount related to the visual guidance field for the extracted attention area, and the layout result evaluation unit evaluates the layout result based on the calculated feature amount. The
As a result, by using the visual guidance field based on physiological and psychological knowledge for layout evaluation, if the evaluation result is reflected in the layout, a relatively good-looking layout can be realized. Can also be obtained.
[0021]
[Invention 6]
Furthermore, the layout evaluation system of the invention 6 is the layout evaluation system of the invention 5,
The feature amount calculation means calculates the strength of the visual guidance field for the attention area extracted by the attention area extraction means, and calculates the guidance field feature value indicating the calculated visual guidance field strength. It is characterized by.
[0022]
With such a configuration, the feature amount calculation means calculates the strength of the visual guidance field for the extracted attention area, and calculates the guidance field feature value indicating the calculated strength of the visual guidance field. The layout result evaluation means evaluates the layout result based on the calculated guidance field feature amount.
As a result, by using the strength of the visual guidance field based on physiological and psychological knowledge for layout evaluation, if the evaluation result is reflected in the layout, a more attractive layout can be realized. The effect is also obtained.
[0023]
[Invention 7]
Furthermore, the layout evaluation system of the invention 7 is the layout evaluation system of any of the inventions 5 and 6,
The feature amount calculating means calculates a visual guidance field for the attention area extracted by the attention area extraction means, obtains an equipotential line from the calculated visual guidance field, and indicates a complexity of the equipotential line The feature amount is calculated.
[0024]
With such a configuration, a visual guidance field is calculated for the extracted attention area by the feature amount calculation means, and an equipotential line is obtained from the calculated visual guidance field, and the complexity of the equipotential line is calculated. Is calculated, and the layout result evaluation unit evaluates the layout result based on the calculated complexity feature amount.
[0025]
By using the complexity of equipotential lines in the visual guidance field based on physiological and psychological knowledge for layout evaluation, if the evaluation result is reflected in the layout, a more attractive layout can be realized. The effect that it can do is also acquired.
[0026]
[Invention 8]
Further, the layout evaluation system of the invention 8 is the layout evaluation system of any of the inventions 5 to 7,
The feature amount calculation means calculates the energy of the visual guidance field for the attention area extracted by the attention area extraction means, and calculates the energy feature amount indicating the calculated visual guidance field energy. It is characterized by.
[0027]
With such a configuration, the feature amount calculation means calculates the visual guidance field energy for the extracted attention area, calculates the calculated guidance field energy indicating the visual guidance field energy, and calculates the layout. The result evaluation means evaluates the layout result based on the calculated guidance field feature.
As a result, by using the visual induction field energy based on physiological and psychological knowledge for layout evaluation, if the evaluation result is reflected in the layout, a more attractive layout can be realized. An effect is also obtained.
[0028]
[Invention 9]
Furthermore, the layout evaluation system of the invention 9 is the layout evaluation system of any of the inventions 3 to 8,
Furthermore, a layout evaluation model storage unit for storing a layout evaluation model in which a relationship between the distribution of the region of interest and the evaluation of the layout result is learned is provided, and the layout result evaluation unit is extracted by the region of interest extraction unit The layout result is evaluated based on the distribution of the attention area and the layout evaluation model stored in the layout evaluation model storage means.
[0029]
In such a configuration, since the relationship between the attention area distribution and the layout result evaluation is learned in the layout evaluation model, the extracted attention area distribution and the layout evaluation model extracted by the layout result evaluation means The layout result is evaluated based on the layout evaluation model of the storage means.
This makes it easy to determine the quality of layout results relatively easily by preparing layout results that take the reader's attention into account into consideration, and letting the layout evaluation model learn the distribution of attention areas for the layout results. The effect that it is possible is also acquired.
[0030]
[Invention 10]
Furthermore, the layout evaluation system of the invention 10 is the layout evaluation system of the invention 9,
The layout evaluation model is a neural network that has learned the relationship between the distribution of the region of interest and the evaluation of the layout result,
The layout result evaluation means inputs a distribution of the attention area extracted by the attention area extraction means to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. And
[0031]
With such a configuration, since the relationship between the attention area distribution and the layout result evaluation is learned in the neural network, the extracted attention area distribution is input to the neural network by the layout result evaluation means. Then, the output value of the neural network is output as an evaluation of the layout result.
[0032]
[Invention 11]
Furthermore, the layout evaluation system of the invention 11 is the layout evaluation system according to any one of the inventions 5 to 8,
Furthermore, a layout evaluation model storage means for storing a layout evaluation model in which a relationship between a feature amount relating to a visual guidance field and an evaluation of the layout result is stored,
The layout result evaluation unit is characterized in that the layout result is evaluated based on the feature amount calculated by the feature amount calculation unit and the layout evaluation model of the layout evaluation model storage unit.
[0033]
With such a configuration, the layout evaluation model has learned the relationship between the feature amount related to the visual induction field and the evaluation of the layout result, so the feature value calculated by the layout result evaluation means and the layout evaluation are evaluated. The layout result is evaluated based on the layout evaluation model of the model storage means.
This makes it easy to see the layout results taking into account the reader's attention, and if the layout evaluation model learns the features related to the visual induction field for each region of interest for the layout results, the quality of the layout results can be improved. The effect that it can determine comparatively easily is also acquired.
[0034]
[Invention 12]
Furthermore, the layout evaluation system of the invention 12 is the layout evaluation system of the invention 11,
The layout evaluation model is a neural network that has learned a relationship between a feature amount relating to a visual induction field and an evaluation of the layout result,
The layout result evaluation unit inputs the feature amount calculated by the feature amount calculation unit to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. .
[0035]
With such a configuration, the neural network has learned the relationship between the feature amount related to the visual guidance field and the evaluation of the layout result, so that the feature value calculated by the layout result evaluation means is transferred to the neural network. The output value of the neural network is output as an evaluation of the layout result.
[0036]
[Invention 13]
Furthermore, the layout evaluation system of the invention 13 is the layout evaluation system of the invention 6,
And further comprising a layout evaluation model storage means for storing a layout evaluation model in which the relationship between the guidance field feature amount indicating the strength of the visual guidance field and the evaluation of the layout result is learned,
The layout result evaluating means is characterized in that the layout result is evaluated based on the guidance field feature quantity calculated by the feature quantity calculating means and the layout evaluation model of the layout evaluation model storage means.
[0037]
With such a configuration, the layout evaluation model has learned the relationship between the guidance field feature amount indicating the strength of the visual guidance field and the evaluation of the layout result, and is calculated by the layout result evaluation means. The layout result is evaluated based on the guidance field feature quantity and the layout evaluation model stored in the layout evaluation model storage means.
In this way, if you prepare an easy-to-read layout result considering the reader's attention location and let the layout evaluation model learn the guidance field feature amount indicating the strength of the visual guidance field for each attention area for the layout result, There is also an effect that the quality of the layout result can be determined relatively easily.
[0038]
[Invention 14]
Furthermore, the layout evaluation system of the invention 14 is the layout evaluation system of the invention 13,
The layout evaluation model is a neural network that has learned the relationship between the induction field feature amount indicating the strength of the visual induction field and the evaluation of the layout result,
The layout result evaluating means inputs the guidance field feature amount calculated by the feature amount calculating means to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. And
[0039]
In such a configuration, the neural network has learned the relationship between the guidance field feature amount indicating the strength of the visual guidance field and the evaluation of the layout result. The guidance field feature is input to the neural network, and the output value of the neural network is output as an evaluation of the layout result.
[0040]
[Invention 15]
Furthermore, the layout evaluation system of the invention 15 is the layout evaluation system of the invention 7,
And further comprising layout evaluation model storage means for storing a layout evaluation model in which a relationship between the complexity feature amount indicating the complexity of the equipotential line and the evaluation of the layout result is learned,
The layout result evaluation means evaluates the layout result based on the complexity feature quantity calculated by the feature quantity calculation means and the layout evaluation model of the layout evaluation model storage means.
[0041]
In such a configuration, the layout evaluation model has learned the relationship between the complexity feature amount indicating the complexity of the equipotential lines in the visual guidance field and the evaluation of the layout result. Thus, the layout result is evaluated based on the calculated complexity feature quantity and the layout evaluation model stored in the layout evaluation model storage means.
[0042]
As a result, an easy-to-read layout result that takes into account the reader's attention location is prepared, and the layout evaluation model learns the complexity feature amount indicating the complexity of the equipotential line in the visual guidance field for each attention area. In this case, it is possible to obtain the effect that the quality of the layout result can be determined relatively easily.
[0043]
[Invention 16]
Furthermore, the layout evaluation system of the invention 16 is the layout evaluation system of the invention 15,
The layout evaluation model is a neural network that has learned the relationship between the complexity feature amount indicating the complexity of the equipotential line and the evaluation of the layout result,
The layout result evaluation unit inputs the complexity feature amount calculated by the feature amount calculation unit to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. And
[0044]
In such a configuration, the neural network has learned the relationship between the complexity feature amount indicating the complexity of the equipotential line in the visual guidance field and the evaluation of the layout result. The calculated complexity feature amount is input to the neural network, and the output value of the neural network is output as an evaluation of the layout result.
[0045]
[Invention 17]
Furthermore, the layout evaluation system of the invention 17 is the layout evaluation system of the invention 8,
And further comprising layout evaluation model storage means for storing a layout evaluation model in which a relationship between an energy feature amount indicating the energy of a visual guidance field and evaluation of the layout result is learned,
The layout result evaluation unit is configured to evaluate the layout result based on the energy feature amount calculated by the feature amount calculation unit and the layout evaluation model of the layout evaluation model storage unit.
[0046]
With such a configuration, the layout evaluation model has learned the relationship between the energy feature amount indicating the energy of the visual induction field and the evaluation of the layout result. The layout result is evaluated based on the feature amount and the layout evaluation model stored in the layout evaluation model storage means.
[0047]
This makes it possible to prepare layout results that are easy to see in consideration of the reader's attention, and for the layout results to learn the energy features indicating the energy of the visual induction field for each region of interest. It is also possible to obtain an effect that it is possible to determine whether the quality is good or bad relatively easily.
[0048]
[Invention 18]
Furthermore, the layout evaluation system of the invention 18 is the layout evaluation system of the invention 17,
The layout evaluation model is a neural network that has learned the relationship between the energy feature amount indicating the energy of the visual induction field and the evaluation of the layout result,
The layout result evaluation means inputs the energy feature amount calculated by the feature amount calculation means to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. To do.
[0049]
In such a configuration, since the relationship between the energy feature amount indicating the energy of the visual induction field and the evaluation of the layout result is learned in the neural network, the energy feature calculated by the layout result evaluation unit is learned. The quantity is input to the neural network, and the output value of the neural network is output as an evaluation of the layout result.
[0050]
[Invention 19]
Furthermore, the layout evaluation system of the invention 19 is the layout evaluation system of any of the inventions 3 to 18,
The layout result evaluating means evaluates the layout result based on the distribution of a predetermined upper number of the plurality of attention areas when the attention area extraction means extracts a plurality of attention areas. It is characterized by that.
[0051]
With such a configuration, when a plurality of attention areas are extracted, the layout result evaluation means evaluates the layout result based on the distribution of the upper predetermined number of the extracted attention areas.
As a result, the layout result is evaluated based on the distribution of the top predetermined number of the plurality of attention areas, and if the evaluation result is reflected in the layout, a layout that is easier to see in consideration of the reader's attention point can be realized. In addition, there is an effect that the processing load can be reduced when there are many attention areas.
[0052]
[Invention 20]
On the other hand, in order to achieve the above object, the layout evaluation program of the invention 20
A program for evaluating the layout,
It is a program for causing a computer to execute processing for evaluating the layout result based on the degree of attraction obtained from the layout result.
With such a configuration, when the program is read by the computer and the computer executes processing according to the read program, the same operation and effect as those of the layout evaluation system of the first aspect can be obtained.
[0053]
[Invention 21]
Furthermore, the layout evaluation program of the invention 21 is
A program for evaluating the layout,
An attention area extraction unit that extracts an attention area of the layout result based on a layout result, and a process that is realized as a layout result evaluation means that evaluates the layout result based on the extraction result of the attention area extraction means. It is a program for making it happen.
With such a configuration, when the program is read by the computer and the computer executes processing according to the read program, the same operation and effect as the layout evaluation system of the invention 3 can be obtained.
[0054]
[Invention 22]
On the other hand, in order to achieve the above object, the layout evaluation method of the invention 22
A method for evaluating a layout,
The layout result is evaluated based on the degree of attraction obtained from the layout result.
Thereby, an effect equivalent to that of the layout evaluation system of aspect 1 can be obtained.
[0055]
[Invention 23]
Furthermore, the layout evaluation method of the invention 23 is
A method for evaluating a layout,
An attention area extracting step of extracting an attention area of the layout result based on a layout result, and a layout result evaluation step of evaluating the layout result based on an extraction result of the attention area extraction step.
[0056]
Thereby, the same effect as the layout evaluation system of the invention 3 is acquired.
Here, the attention area extraction step may extract the attention area of the layout result based on the layout result. For example, the degree of attraction can be calculated based on the layout result, and the attention area of the layout result can be extracted based on the calculated degree of attraction.
[0057]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIGS. 1 to 12 are diagrams showing a first embodiment of a layout evaluation system, a layout evaluation program, and a layout evaluation method according to the present invention.
In the present embodiment, the layout evaluation system, the layout evaluation program, and the layout evaluation method according to the present invention are learned in advance in the neural network 300 for easy-to-read layout results that take into account the reader's attention, and the neural network 300 is used. This is applied to the case of evaluating a given layout result.
[0058]
In the present embodiment, the concept of “attraction level” is used as a reference for extracting a portion (hereinafter referred to as a region of interest) that the reader is likely to focus on in the layout result. A method for calculating the degree of attraction is disclosed in detail in, for example, “Japanese Patent Application Laid-Open No. 2001-127070 (attention area extraction apparatus and automatic composition determination apparatus using the same).
Briefly explain the degree of attraction.
[0059]
In order to extract a region of interest, the degree of attraction is evaluated according to the physical characteristics of the original image. Here, the degree of attraction means a parameter that matches the human subjectivity. In the extraction of the attention area, the area that stands out from the evaluation result is extracted as the attention area. That is, in the evaluation of the attention area, since the evaluation according to the human subjectivity is performed according to the physical characteristics, the attention area suitable for the human subjectivity can be extracted.
[0060]
For example, when the physical feature includes a degree of color heterogeneity, the degree of attraction can be evaluated based on the difference in color of each region.
In addition to the color heterogeneity, the physical characteristics further include a shape heterogeneity, an area heterogeneity, and a texture (pattern) heterogeneity, and therefore based on at least one of these four heterogeneities. If the degree of attraction is evaluated, the degree of attraction can be accurately evaluated according to the characteristics of the original image.
[0061]
In the case of evaluating the three elements of color (hue, saturation and lightness), an area close to a conspicuous color (red) according to human subjectivity can be evaluated as the most prominent area.
Further, if the spatial frequency and the area of each region in the original image are also evaluated, the most conspicuous region can be evaluated more accurately.
[0062]
The degree of attraction is evaluated according to the following procedure.
(1) First, the original image is divided into regions. In this case, the original image is divided into a figure area and a picture area. The boundary detection method based on “edge flow” described in “Edge Flow: A Framework of Boundary Detection and Image Segmentation” by WYMa and BSManjunath et al. In 1997 IEEE is applied.
(2) Next, the divided figure area is extracted, and the attractiveness of the area is evaluated.
[0063]
The degree of attraction is evaluated as follows.
First, the degree of heterogeneity attraction in each area is obtained. In this case, the heterogeneity of the color, the heterogeneity of the texture, the heterogeneity of the shape, and the heterogeneity of the area are obtained, and a weight coefficient is assigned to each to linearly combine to obtain the heterogeneity attraction of each region.
Next, the degree of feature attraction in each area is obtained. In this case, the attractiveness of the color, the attractiveness of the spatial frequency, and the attractiveness of the area are obtained, and a weighting coefficient is assigned to each to linearly combine to obtain the feature attractiveness of each region.
[0064]
Next, the heterogeneity attractiveness degree and the feature attractiveness degree of each region are added to obtain a feature amount integrated value, and the feature amount integrated value is evaluated by a predetermined beta function to calculate the attractiveness degree.
(3) Moreover, the pattern figure which evaluated the attractiveness degree from the original image is produced | generated.
In this embodiment, the concept of “visual guidance field” is used as an evaluation criterion for layout results.
[0065]
Briefly explain the visual induction field.
The visual guidance field is used, for example, as an index of readability of the entire character string by evaluating the readability of individual characters existing on the character string.
First, as an example of estimating a visual induction field of a character image based on physiological and psychological knowledge, a method of estimating a visual induction field from a digital image of a character obtained by digitization will be described.
[0066]
In addition, it is said that the state where each character in the character string is easy to read is that visual guidance fields surrounding the individual characters are arranged at intervals so as not to interfere with each other as much as possible. Specifically, when a closed curve of a visual induction field surrounding each character is considered, if the potential value of the closed curve is high, it is difficult to separate from other characters and difficult to read. From this, it is considered that the readability of each character in the character string can be quantitatively evaluated based on the spread of the visual induction field. The visual guidance field is described in Yoshimasa Yokose's “Psychology of Shape” (Nagoya University Press (1986)) (hereinafter referred to as a reference paper).
[0067]
The visual guidance field shown in the reference paper (hereinafter simply abbreviated as guidance field) explains the visual phenomenon by considering the “field” that spreads around the figure. Since the reference paper targets figures composed of straight lines and arcs, an arbitrary digital image induction field is not required. Here, a method for calculating the induction field in a black and white binary digital image is shown first.
[0068]
Since the induction field can be basically interpreted as a Coulomb potential, the pixels constituting the outline of the pattern are assumed to be point charges, and the distribution of the induction field in the digital image is calculated from the accumulation of the Coulomb potential created by them.
FIG. 1 is a diagram showing a pixel arrangement of a digital image.
As shown in FIG. 1, it is assumed that an induction field is formed at an arbitrary point P by a curve f (s) composed of n point sequences. The curve f (s) hits the line segment of the line figure and the outline of the picture figure. And each point p constituting the curve f (s) ₁ , P ₂ , ..., p _i , ..., p _n Is a point charge of positive charge 1, and the point f is scanned from the point P on the curve f (s), and n points p constituting the curve f (s) are obtained. ₁ , P ₂ , ..., p _i , ..., p _n And the distance to each point on the curve f (s) found by scanning _i Then, the strength M of the induction field at point P _xy Can be defined by the following equation (1). M _xy The subscript xy represents the x coordinate and the y coordinate in the image of the point P.
[0069]
[Expression 1]

[0070]
FIG. 2 is a diagram for explaining a shielding condition for obtaining the strength of the induction field.
By using the above equation (1), the induction field of an arbitrary digital image can be obtained. When there are a plurality of curves, the strength M of the induction field at the point P _xy Is the sum of the induction fields created by individual curves at point P. Note that the above equation (1) has a constraint that a sum is obtained only for a portion directly irradiated with light emitted from the point P. For example, for point P, curve f ₁ (s), f ₂ (s), f _Three If (s) exists as shown in FIG. 2, the portion that cannot be seen from the point P, that is, in this case, the curve f ₁ The sum of the portions in the range Z that is shielded by (s) and cannot be seen from the point P is not taken. In the example of FIG. 2, the curve f _Three All of (s) and curve f ₂ Some sums of (s) will not be taken. This is called a shielding condition.
[0071]
FIG. 3 shows an example of the guidance field of the letter “A”. FIG. 3A shows the guidance field when the guidance field is calculated in consideration of the shielding condition, and FIG. It is a figure which shows the case where it calculated | required.
FIG. 3A shows an example of the induction field calculated by the above formula (1) for the letter “A”. A thin line L distributed in contour lines on the map around the letter “A” in FIG. 3A is an equipotential line of the induction field, and the intensity M of the induction field increases from the center to the outside. _xy Will soon become weak and will approach zero.
[0072]
The characteristics in the shape and strength of the induction field distribution shown in FIG. 3A, in particular, the distribution near the apex of “A” is sharper than others. It agrees with the psychological experiment result of the distribution of.
FIG. 3B is an example of an induction field that has no shielding condition and assumes that all pixels are point charges with a positive charge of 1. However, the distribution of the induction field is rounded as a whole, and the psychology according to the reference paper. The result is different from the experimental result. Thus, the shielding condition is important in characterizing the induction field.
[0073]
In this way, a guidance field for a certain character can be obtained. Examples of techniques using guidance fields include, for example, “Michihiro Nagaishi:“ Easy-to-read Japanese proportional display using visual guidance fields ”, Journal of the Institute of Image Media and Technology, Vol.52, No.12, pp.1865. -1872 (1998) "(hereinafter referred to as the first paper)," Masayoshi Miyoshi, Yoshifumi Shimoshiro, Hiroaki Koga, Ken Ideguchi: "Design of character layout based on sensibility using visual induction field theory", IEICE Transactions, 82-A, 9, 1465-1473 (1999) ”(hereinafter referred to as the second paper).
[0074]
In this embodiment, when such a guidance field is used to display a group of images made up of characters, photos, pictures, figures, etc. within a certain predetermined display range, the layout is optimal. Evaluate the layout so that
In this embodiment, when evaluating whether the layout is good or bad, a group of images to be laid out is regarded as one induction field calculation object, the induction field is calculated, and the equipotential lines obtained thereby are calculated. Evaluate the quality of the layout based on the shape.
[0075]
FIG. 4 is a diagram showing an image of a part of a newspaper article as a reference layout example.
Now, it is assumed that a group of images to be laid out is an image composed of a character string and a photograph as shown in FIG. The image shown in FIG. 4 shows a part of a newspaper article, and consists of a character string portion C and photographs P1 and P2, and the layout shown in FIG. 4 is made by a designer specialized in newspapers. Suppose the layout is easy for many people to see and understand.
[0076]
FIG. 5 shows the image shown in FIG. 4, where the character string part represents each character string as a simple line, and the photograph was simply represented by a rectangular frame to calculate the induction field, and was obtained from the calculated induction field. It is a figure which shows an equipotential line.
As shown in FIG. 4, when the induction field is calculated using the above equation (1) for the entire group of images laid out in a limited display range, as shown in FIG. An equipotential line L is drawn. When the guidance field is calculated for the entire information to be laid out, the character string portion C shown in FIG. 4 represents each character string as a simple line as shown in FIG. , P2 expresses the outer shape with a rectangular frame and calculates the induction field.
[0077]
This is because the layout is determined by the positional relationship and size of each element, so each element can be expressed in a simplified manner. In this way, the induction field is calculated with each element expressed in a simplified manner. If an equipotential line is drawn from the obtained induction field, the equipotential line can represent the equipotential line of the entire layout.
[0078]
Note that the layout shown in FIG. 4 is a layout designed by a professional designer that is easy to see and understand the contents. The equipotential lines L obtained from the entire image thus laid out are There are few irregularities and it is rounded.
From this, it is possible to determine the quality of the layout of the image from the shape of the equipotential line obtained by calculating the induction field for the entire group of images to be laid out. That is, if the degree of unevenness of the obtained equipotential lines is known, it can be evaluated whether the layout of the image is a good layout.
[0079]
Therefore, in the present embodiment, the degree of unevenness of the equipotential line is obtained as the complexity of the equipotential line, and the complexity is used as an index for evaluating the quality of the layout of the image. In other words, the more the equipotential line is rounded with less unevenness, the smaller the complexity, and the more the equipotential line is uneven, the greater the complexity. Complexity is the complexity of the i-th equipotential line, C _i It can be defined by the following formula (2). In the above equation (2), L _i Is the length of the i-th equipotential line, S _i Represents the area of the surface surrounded by the i-th equipotential line. Note that the length L of the i-th equipotential line _i Can be considered as the number of dots constituting the potential line, and the area S of the surface surrounded by the i-th equipotential line _i Can be considered as the number of dots existing on the surface surrounded by the i-th equipotential line.
[0080]
[Expression 2]

[0081]
According to the above equation (2), as the length of the equipotential line drawn by the induction field calculated for the group of images to be laid out becomes longer (the more unevenness), the complexity C _i It can be said that the value of increases. In other words, the equipotential line has fewer irregularities and is closer to a circle, the more complex C _i Is a small value.
FIG. 6 is a diagram in the case where the reference layout shown in FIG. 4 and a layout in which the reference layout is variously changed are used.
[0082]
Now, let us calculate the complexity of the group of images shown in FIG. 4 when they are arranged in various layouts as shown in FIG. In FIG. 6, as in FIG. 5, the character string portion C represents each character string by a simple line, and the photographs P1 and P2 are simply represented by a rectangular frame.
6A is the same layout as FIG. 4 (this is referred to as layout A1), and FIG. 6B is a layout in which the photo P2 of FIG. (C) is a layout in which the photograph P1 is in the lower right and the photograph P2 is in the upper left (this is referred to as layout A3), and (d) in FIG. This is a layout in which P2 is arranged in a character string (this is called layout A4).
[0083]
FIG. 7 is a diagram showing the complexity of each layout when the layouts shown in FIGS.
For these, first, the respective induction fields are calculated, and the complexity is calculated by the above equation (2) from the equipotential lines (each i-th potential line) drawn by the obtained induction fields. A result like 7 was obtained. In FIG. 7, the horizontal axis represents the respective layouts A1 to A6, and the vertical axis represents the complexity obtained for each of the layouts A1 to A6.
[0084]
According to FIG. 7, the layout A1 (referred to as reference layout A1), which is laid out by the designer and is easy to read and understand the contents, has the smallest complexity, and the other three layouts A2, A3, A4 However, compared with the standard layout A1, its complexity is a large value. In particular, in this example, the layout A3 has the greatest complexity.
[0085]
This is because, as described above, the induction field obtained from the reference layout A1 has less unevenness and is generally rounded, and the other three layouts A2 to A4 are obtained from the respective layouts, etc. This is because the potential line has large irregularities.
Further, by using equipotential lines, the energy E of the induction field in the entire image can be defined by the following equation (3). In the following formula (3), i represents the i-th equipotential line, S _i Is the area of the surface surrounded by the i-th equipotential line, P _i Represents the potential value in the i-th equipotential line. This corresponds to obtaining the volume of the induction field when the induction field is considered three-dimensionally, and the size of the volume is defined as energy.
[0086]
[Equation 3]

[0087]
The above is a case where a part of an article such as a newspaper (mostly composed of character strings and photos) is set as a group of images to be laid out and evaluation is performed on the layout of the group of images. As a layout target image, an evaluation using a general image can be considered similarly.
Next, the configuration of the layout evaluation apparatus according to the present invention will be described with reference to FIG.
[0088]
FIG. 8 is a functional block diagram showing the configuration of the layout evaluation apparatus according to the present invention.
As shown in FIG. 8, the layout evaluation apparatus according to the present invention includes an evaluation unit 100 that evaluates a layout result and a learning unit 200 that learns an easy-to-see layout result in consideration of a reader's attention location. Specifically, it is configured as a general computer in which a CPU, ROM, RAM, I / F, and the like are connected by bus, and the CPU activates a predetermined program stored in a predetermined area of the ROM and evaluates according to the program. The processing realized as the unit 100 and the learning unit 200 is executed.
[0089]
The evaluation unit 100 extracts the layout result storage unit 110 that stores the layout result, the attention region extraction unit 115 that extracts the attention region of the layout result based on the layout result of the layout result storage unit 110, and the attention region extraction unit 115. A feature amount calculation unit 120 that calculates a feature amount related to the guidance field for the attention region, a attention region distribution model storage unit 130 that stores a relationship between the distribution of the attention region and the evaluation of the layout result as a attention region distribution model, The layout result evaluation unit 140 evaluates the layout result of the layout result storage unit 110 based on the feature amount calculated by the calculation unit 120 and the attention region distribution model of the attention region distribution model storage unit 130.
[0090]
The attention area extraction unit 115 reads out the layout result to be evaluated from the layout result storage unit 110, calculates the degree of attraction based on the read layout result, and extracts the attention area based on the calculated degree of attraction. ing. The region of interest is extracted by the above method. When a plurality of attention areas are extracted, the top N attention areas (N is a natural number) having a high degree of attraction are extracted. Hereinafter, the degree of attraction calculated for each pixel constituting the layout result is F (x, y). x and y indicate the X coordinate and Y coordinate of the pixel in the layout result.
[0091]
The feature amount calculation unit 120 applies the guidance field strength M for the attention region extracted by the attention region extraction unit 115. _xy , Complexity of equipotential lines C _i In addition, the energy E of the induction field is calculated as a feature amount. Induction field strength M _xy , Complexity of equipotential lines C _i The energy E of the induction field is calculated based on an image obtained by performing a black and white binarization process on the layout result. In the present embodiment, the degree of attractiveness F (x, y) and each feature amount M _xy , C _i And E are treated as vectors.
[0092]
FIG. 9 is a diagram illustrating a configuration of the neural network 300.
The attention area distribution model storage unit 130 stores the distribution of the attention area by the neural network 300 as shown in FIG.
As shown in FIG. 9, the neural network 300 includes an attractiveness degree F (x, y) and a feature amount M. _xy , C _i And i input layers I for inputting E and E _i And each input layer I _i J intermediate layers H to which outputs from _j And each intermediate layer H _j Output layer O that outputs the evaluation value by inputting the output of _k It consists of and. And the input layer I _i And middle layer H _j Is the coupling coefficient W _ij The middle layer by synapse _j And output layer O _k Is the coupling coefficient W _jk Are connected by synapses.
[0093]
In addition, the neural network 300 learns features of a layout result that is easy to see in consideration of a reader's attention by a feature learning unit 230 described later. Therefore, the attractiveness F (x, y) and the feature amount M calculated from the easy-to-see layout result in consideration of the reader's attention location. _xy , C _i When E and E are input to the neural network 300, a relatively high value is obtained as the evaluation value. _k The degree of attractiveness F (x, y) and the feature quantity M calculated from the layout result that does not take into account the reader's attention point _xy , C _i When E and E are input to the neural network 300, the evaluation layer has a relatively low value as the output layer O. _k Is output from.
[0094]
Returning to FIG. 8, the layout result evaluation unit 140 includes the attractiveness F (x, y) calculated by the attention area extraction unit 115 and the feature amount M calculated by the feature amount calculation unit 120. _xy , C _i And E are input to the neural network 300, and the output value of the neural network 300 is output as an evaluation of the layout result. The output evaluation value may be displayed on a display device such as a display, or may be used for selecting a layout result or determining a layout.
[0095]
As illustrated in FIG. 8, the learning unit 200 includes a layout result designation input unit 210 that inputs designation of a layout result to be learned from the layout result storage unit 110, and a layout among layout results of the layout result storage unit 110. A feature region extraction unit 215 that extracts a region of interest in a layout result based on the designation input by the result designation input unit 210, and a feature that calculates a feature amount related to the guidance field for the region of interest extracted by the region of interest extraction unit 215 An amount calculation unit 220 and a feature learning unit 230 that learns features of a layout result that is easy to see based on the feature amount calculated by the feature amount calculation unit 220 in consideration of a reader's attention location.
[0096]
The attention area extraction unit 215 is configured to have the same function as the attention area extraction unit 115, and is based on the layout result of the layout result storage unit 110 related to the designation input by the layout result designation input unit 210. The degree of attraction F (x, y) is calculated, and the attention area is extracted based on the calculated degree of attraction F (x, y).
[0097]
The feature amount calculation unit 220 is configured to have the same function as the feature amount calculation unit 120, and the guidance field strength M for the attention region extracted by the attention region extraction unit 115. _xy , Complexity of equipotential lines C _i In addition, the energy E of the induction field is calculated as a feature amount.
The feature learning unit 230 includes an attractiveness degree F (x, y) calculated by the attention area extraction unit 215 and a feature amount M calculated by the feature amount calculation unit 220. _xy , C _i On the basis of E and E, the neural network 300 is learned by a known back-propagation method or other learning methods. In learning, the degree of attraction F (x, y) and the feature amount M calculated from the layout result related to the designation input by the layout result designation input unit 210. _xy , C _i When E and E are input to the neural network 300, the output layer O is relatively high as the evaluation value. _k Coupling coefficient W to be output from _ij , W _jk To decide. For example, when the back propagation method is used, the coupling coefficient W is determined by forward calculation or backward calculation. _ij , W _jk To decide.
[0098]
Next, the operation of the present embodiment will be described.
First, a case where the neural network 300 is learned will be described.
When learning the neural network 300, the user or designer designates several layout results that are easy to see from the layout result storage unit 110 in consideration of the reader's attention. This designation is input to the layout result designation input unit 210.
[0099]
When the specification of the layout result is input, the attention area extraction unit 215 calculates the attractiveness F (x, y) based on the input specification related to the input layout result of the layout result storage unit 110, and calculates A region of interest is extracted based on the degree of attraction F (x, y). Next, the intensity M of the induction field for the extracted attention area by the feature amount calculation unit 220. _xy , Complexity of equipotential lines C _i And the energy E of the induction field is calculated as a feature amount. Then, the feature learning unit 230 calculates the attractiveness degree F (x, y) and the feature amount M calculated. _xy , C _i The neural network 300 is learned based on E and E. This series of processing is performed for all designated layout results.
[0100]
Next, a case where a layout is evaluated will be described.
FIG. 10 is a diagram showing images 600 and 602 to be laid out.
FIG. 11 is a diagram showing layout results 610 and 620 to be evaluated.
Now, as shown in FIG. 10, there are two images 600 and 602 to be laid out. Two layouts were created with different layout patterns using these two images. The layout results are as shown in FIGS. 10 (a) and 10 (b), respectively. In the layout result 610, as shown in FIG. 10A, an image 600 is arranged in the upper left area and an image 602 is arranged in the lower right area, and the sizes of the images 600 and 602 are almost the same. Sentences corresponding to the images 600 and 602 are arranged in the upper right area and the lower left area. In addition, as shown in FIG. 10B, the layout result 620 is an image in which an image 600 is arranged in the upper left area, an image 602 is arranged in the lower right area, and the image 600 is larger than the image 602 and is arranged mainly. It is. Sentences corresponding to the images 600 and 602 are arranged in the upper right area and the lower left area.
[0101]
FIG. 12 is a diagram illustrating a region of interest of the layout results 610 and 620.
First, the attention area extraction unit 115 reads the layout result 610 from the layout result storage unit 110, calculates the attractiveness F (x, y) based on the read layout result 610, and calculates the attractiveness calculated. A region of interest is extracted based on F (x, y). When a plurality of attention areas are extracted, the top N (for example, four) attention areas having a high degree of attraction F (x, y) are extracted. As shown in FIG. 12A, the attention area of the layout result 610 is extracted as a hatched area in the upper left area, the upper right area, the lower left area, and the lower right area. For convenience of explanation, the attention area is indicated by an ellipse, but in reality, the shape of the area changes depending on the method of area division.
[0102]
Next, the feature amount calculation unit 120 performs guidance field strength M for each region of interest in the layout result 610. _xy , Complexity of equipotential lines C _i And the energy E of the induction field is calculated as a feature amount. Then, the attraction degree F (x, y) calculated by the layout result evaluation unit 140 and the feature amount M are calculated. _xy , C _i And E are input to the neural network 300, and the output value of the neural network 300 is output as an evaluation of the layout result.
[0103]
Similarly, the attention area extraction unit 115 reads the layout result 620 from the layout result storage unit 110, calculates the degree of attraction F (x, y) based on the read layout result 620, and calculates the attraction calculated A region of interest is extracted based on the degree F (x, y). When a plurality of attention areas are extracted, the top N attention areas having a high degree of attraction F (x, y) are extracted. As shown in FIG. 12B, the attention area of the layout result 620 is extracted as a hatched area in the upper left area, upper right area, lower left area, and lower right area. The difference from the layout result 610 is that the attention area corresponding to the image 600 is large and the attention area corresponding to the image 602 is small.
[0104]
Next, the feature amount calculation unit 120 performs guidance field strength M for each region of interest in the layout result 620. _xy , Complexity of equipotential lines C _i And the energy E of the induction field is calculated as a feature amount. Then, the attraction degree F (x, y) calculated by the layout result evaluation unit 140 and the feature amount M are calculated. _xy , C _i And E are input to the neural network 300, and the output value of the neural network 300 is output as an evaluation of the layout result.
[0105]
Thus, in this embodiment, the attention area extraction unit 115 that extracts the attention area of the layout result based on the layout result, and the layout result is evaluated based on the distribution of the attention area extracted by the attention area extraction section 115. A layout result evaluation unit 140.
As a result, the layout result is evaluated based on the attention area. Therefore, if the evaluation result is reflected in the layout, it is possible to realize an easy-to-view layout in consideration of the reader's attention position as compared with the conventional case.
[0106]
Furthermore, in the present embodiment, the feature amount calculation unit 120 that calculates the feature amount related to the guidance field for the attention region extracted by the attention region extraction unit 115 is provided, and the layout result evaluation unit 140 calculates the feature amount by the feature amount calculation unit 120. The layout result is evaluated based on the feature amount.
Thus, by using the guidance field based on physiological and psychological knowledge for layout evaluation, if the evaluation result is reflected in the layout, a relatively good-looking layout can be realized.
[0107]
Furthermore, in the present embodiment, the neural network 300 is made to learn the relationship between the degree of attraction F (x, y) and the layout result evaluation, and the layout result evaluation unit 140 calculates the attraction calculated by the attention area extraction unit 115. The degree F (x, y) is input to the neural network 300, and the output value of the neural network 300 is output as an evaluation of the layout result.
[0108]
As a result, an easy-to-read layout result considering the reader's attention location is prepared, and if the neural network 300 learns the distribution of the attractiveness F (x, y) for the layout result, the quality of the layout result is relatively good. It can be easily determined.
Furthermore, in this embodiment, the strength M of the induction field _xy The neural network 300 learns the relationship between the feature quantity indicating the layout result and the evaluation of the layout result, and the feature quantity calculation unit 120 uses the guidance field strength M for the attention area extracted by the attention area extraction section 115. _xy The layout result evaluation unit 140 calculates the strength M of the induction field calculated by the feature amount calculation unit 120. _xy Is input to the neural network 300, and the output value of the neural network 300 is output as an evaluation of the layout result.
[0109]
As a result, the strength of the induction field based on physiological and psychological knowledge M _xy Since the evaluation result is reflected in the layout, a more attractive layout can be realized. Also, an easy-to-read layout result that takes into account the reader's attention location is prepared, and the strength of the induction field M for each attention area for the layout result. _xy If the neural network 300 is made to learn the feature quantity indicating “”, the quality of the layout result can be determined relatively easily.
[0110]
Further, in the present embodiment, the equipotential line complexity C _i , The neural network 300 learns the relationship between the feature amount indicating the layout result and the layout result evaluation, and the feature amount calculation unit 120 calculates a guidance field for the attention region extracted by the attention region extraction unit 115, and the calculated guidance field. To obtain an equipotential line from the complexity of the equipotential line C _i The layout result evaluation unit 140 calculates the complexity C calculated by the feature amount calculation unit 120. _i Is input to the neural network 300, and the output value of the neural network 300 is output as an evaluation of the layout result.
[0111]
As a result, the complexity C of the equipotential line in the induction field based on physiological and psychological knowledge _i Since the evaluation result is reflected in the layout, a more attractive layout can be realized. In addition, an easy-to-read layout result considering the reader's attention location is prepared, and the equipotential line complexity C for each attention area in the layout result is prepared. _i If the neural network 300 is made to learn the feature quantity indicating “”, the quality of the layout result can be determined relatively easily.
[0112]
Furthermore, in the present embodiment, the neural network 300 is made to learn the relationship between the feature amount indicating the energy E of the induction field and the evaluation of the layout result, and the feature amount calculation unit 120 is extracted by the attention area extraction unit 115. The guidance field energy E is calculated for the region of interest, and the layout result evaluation unit 140 inputs the guidance field energy E calculated by the feature amount calculation unit 120 to the neural network 300 and outputs the output value of the neural network 300 as the layout result. Output as an evaluation.
[0113]
As a result, by using the energy E of the induction field based on physiological and psychological knowledge for layout evaluation, if the evaluation result is reflected in the layout, a more attractive layout can be realized. In addition, if a layout result that is easy to see in consideration of the reader's attention location is prepared and the neural network 300 learns the feature quantity indicating the induction field energy E for each attention area, the layout result is good or bad. Can be determined relatively easily.
[0114]
Further, in the present embodiment, when the attention area extraction unit 115 extracts a plurality of attention areas, the layout result evaluation section 140 evaluates the layout result based on the distribution of the top N of the plurality of attention areas. It is supposed to be.
As a result, the layout result is evaluated based on the distribution of the top N of the plurality of attention areas, and if the evaluation result is reflected in the layout, a layout that is easier to see in consideration of the reader's attention point can be realized. In addition, the processing load can be reduced when there are many attention regions.
[0115]
In the first embodiment, the attention area extraction unit 115 corresponds to the attention area extraction means of the invention 3 to 10, 19 or 21, or the attraction degree calculation means of the invention 2, and the extraction by the attention area extraction section 115 Corresponding to the attention area extracting step of the invention 23, the feature amount calculating unit 120 corresponds to the feature amount calculating means of the inventions 5 to 8 and 11 to 18. The attention area distribution model storage unit 130 corresponds to the layout evaluation model storage unit of the invention 9, 11, 13, 15 or 17, and the layout result evaluation unit 140 is the layout of the invention 2 to 5, 9 to 19 or 21. Corresponding to the result evaluation means, the evaluation by the layout result evaluation unit 140 corresponds to the layout result evaluation step of the invention 23.
[0116]
In the first embodiment, the neural network 300 corresponds to the layout evaluation models of the inventions 9 to 18.
Next, a second embodiment of the present invention will be described with reference to the drawings. FIG. 13 is a diagram showing a second embodiment of a layout evaluation system, a layout evaluation program, and a layout evaluation method according to the present invention.
[0117]
In the present embodiment, the layout evaluation system, the layout evaluation program, and the layout evaluation method according to the present invention are applied to a case where a layout result is evaluated in consideration of a reader's attention point. The point of difference is that the layout result is evaluated based on the attractiveness degree F (x, y) calculated by the attention area extraction unit 115. Hereinafter, only the parts different from the first embodiment will be described, and the same parts as those in the first embodiment will be denoted by the same reference numerals and the description thereof will be omitted.
[0118]
Next, the configuration of the layout evaluation apparatus according to the present invention will be described with reference to FIG.
FIG. 13 is a functional block diagram showing the configuration of the layout evaluation apparatus according to the present invention.
As shown in FIG. 13, the layout evaluation apparatus according to the present invention lays out based on the attractiveness F (x, y) calculated by the layout result storage unit 110, the attention area extraction section 115, and the attention area extraction section 115. The layout result evaluation unit 150 is configured to evaluate the result.
[0119]
The layout result evaluation unit 150 weights the attractiveness F (x, y) extracted by the attention area extraction unit 115 by the Gaussian function G (x, y) by the following formula (4), and calculates the layout evaluation value H. Then, the calculated layout evaluation value H is output as an evaluation of the layout result.
[0120]
[Expression 4]

[0121]
Next, the operation of the present embodiment will be described.
First, the attention area extraction unit 115 reads the layout result 610 from the layout result storage unit 110, calculates the attractiveness F (x, y) based on the read layout result 610, and calculates the attractiveness calculated. A region of interest is extracted based on F (x, y). Then, the layout result evaluation unit 150 calculates the layout evaluation value H by the above equation (4), and the calculated layout evaluation value H is output as the evaluation of the layout result.
[0122]
In this way, in this embodiment, the attention area extraction unit 115 that calculates the attractiveness degree F (x, y) based on the layout result, and the attractiveness degree F (x, y) calculated by the attention area extraction section 115. A layout result evaluation unit 150 that evaluates the layout result based on
Thereby, since the layout result is evaluated based on the degree of attraction, if the evaluation result is reflected in the layout, it is possible to realize an easy-to-view layout that takes into account the reader's attention as compared with the conventional case.
[0123]
In the second embodiment, the attention area extraction unit 115 corresponds to the attention area extraction means of the invention 3, 4 or 21, or the attraction degree calculation means of the invention 2, and the extraction by the attention area extraction section 115 is an invention. The layout result evaluation unit 150 corresponds to the layout result evaluation unit 150 according to the second to fourth or twenty-first aspects. The evaluation by the layout result evaluation unit 150 corresponds to the layout result evaluation step of the twenty-third aspect.
[0124]
In the first embodiment, the layout result evaluation unit 140 uses the attractiveness degree F (x, y) calculated by the attention area extraction unit 115 and the feature amount M calculated by the feature amount calculation unit 120. _xy , C _i And E are input to the neural network 300, and the output value of the neural network 300 is output as the evaluation of the layout result. However, the layout result evaluation unit 140 is not limited to this, and does not use the neural network 300. Attraction degree F (x, y) calculated by the region extraction unit 115 and feature amount M calculated by the feature amount calculation unit 120 _xy , C _i A layout result can be evaluated based on E and E. For example, an evaluation value can be calculated by performing a predetermined weighting on the absolute value of each value and adding the result, and the layout result can be evaluated based on the magnitude of the evaluation value.
[0125]
In the first embodiment, when learning the neural network 300, an easy-to-see layout result considering the reader's attention point is selected from the layout result storage unit 110. The layout result may be selected from the layout result storage unit 110 without being limited to the layout result. In this case, it is sufficient that a plurality of users or designers specify layout results that are easy to see, and the neural network 300 learns the characteristics of the specified layout results in the same manner as in the first embodiment. .
[0126]
Furthermore, in this case, it is possible not only to have a plurality of users or designers input good / bad impressions but also to input strong / weak impressions, and to make the neural network 300 learn based on this. is there. Thereby, since general evaluation can be learned, layout evaluation suitable for a plurality of people's preferences can be performed.
[0127]
Furthermore, in this case, for example, teens, 20s, 30s, etc., categorize according to age and ask each group to specify the layout result that the user or designer thinks is good impression. It is also possible to cause the neural network 300 to learn the characteristics of the layout result. Thereby, the layout evaluation suitable for the preference of the same generation person can be performed. It can also be used to find out what generation of people prefer a layout result.
[0128]
In the first embodiment, the neural network 300 includes the output layer O. _k However, the present invention is not limited to this, and a plurality of output layers may be provided. For example, a first output layer that outputs one of the user's likes / dislikes, a second output layer that outputs one of the user's impressions of good / bad, and the strength / weakness of the user's impression A third output layer that outputs either of them may be provided.
[0129]
In the first embodiment, the feature amount calculation units 120 and 220 are configured to calculate the feature amounts of all the pixels constituting the attention area and perform learning based on the calculated feature amounts. However, the present invention is not limited to this, and among the pixels constituting the region of interest, for example, in a rectangular pixel group consisting of five pixels in the vertical direction and five pixels in the horizontal direction, four corner pixels are targeted, and feature amounts of the target pixel (for example, , An average value) may be calculated, and learning may be performed based on the calculated feature amount.
[0130]
In the first embodiment, the attractiveness F (x, y) calculated for the attention area and the feature amount M are calculated. _xy , C _i However, the present invention is not limited to this, and the degree of attraction F (x, y) and the feature amount M calculated for the attention area are not limited thereto. _xy , C _i And E can be configured to perform layout evaluation and learning based on one of E and E.
[0131]
In the first embodiment, the layout result evaluation unit 140 uses the attractiveness F (x, y) calculated for the attention area and the feature amount M. _xy , C _i However, the present invention is not limited to this, and the layout result may be evaluated based on the total area of the region of interest.
In the first embodiment, the back-propagation method is exemplified as the learning method of the neural network 300. However, the present invention is not limited to this, and an unsupervised learning method by self-organization can also be used. Thereby, for example, the features of 25 layout results created by the user can be learned, and learned along the trend of the layout results, and the user's preferences can be automatically learned.
[0132]
Further, in the first embodiment, the intensity M of the induction field is based on the image obtained by performing the binarization processing on the layout result. _xy , Complexity of equipotential lines C _i However, the present invention is not limited to this, and the intensity M of the induction field is determined based on the color layout result itself. _xy , Complexity of equipotential lines C _i It can also be configured to calculate the energy E of the induction field.
[0133]
In the first embodiment, the neural network 300 is used. Alternatively, a probabilistic model such as “Naive Bayes or Bayesian Network” may be used. Even with such a configuration, an effect equivalent to that of the first embodiment can be obtained.
In the second embodiment, the attention area extraction unit 115 is configured to calculate the attractiveness F (x, y) of the attention area without performing weighting. It is also possible to calculate the attractiveness F (x, y) of the attention area by performing weighting that decreases from the center to the outside.
[0134]
In the second embodiment, the layout result evaluation unit 150 uses the Gaussian function G (x, y) as the degree of attraction F (x, y) extracted by the attention area extraction unit 115 according to the above equation (4). The layout evaluation value H is calculated, and the calculated layout evaluation value H is output as the evaluation of the layout result. However, instead of the Gaussian function G (x, y), linear interpolation or sigmoid function is used. Can be used.
[0135]
In the first embodiment, the layout evaluation system, the layout evaluation program, and the layout evaluation method according to the present invention have the neural network 300 learn in advance a layout result that is easy to see in consideration of the reader's attention. The present invention is applied to the case where a given layout result is evaluated using the neural network 300. However, the present invention is not limited to this and can be applied to other cases without departing from the gist of the present invention.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a pixel arrangement of a digital image.
FIG. 2 is a diagram illustrating a shielding condition when obtaining the strength of an induction field.
FIG. 3 is an example of the guidance field of the letter “A”. FIG. 3A shows the guidance field when the guidance field is obtained in consideration of the shielding condition, and FIG. It is a figure which shows the case where it calculated | required.
FIG. 4 is a view showing an image of a part of a newspaper article as a reference layout example;
FIG. 5 is obtained from the calculated induction field by calculating the induction field by expressing the character string portion with a simple line in the image shown in FIG. It is a figure which shows an equipotential line.
6 is a diagram in the case where the reference layout shown in FIG. 4 and a layout in which the reference layout is variously changed are used.
7 is a diagram showing the complexity of each layout when the layouts shown in FIGS. 6 (a) to 6 (d) are used. FIG.
FIG. 8 is a functional block diagram showing a configuration of a layout evaluation apparatus according to the present invention.
9 is a diagram showing a configuration of a neural network 300. FIG.
FIG. 10 is a diagram showing images 600 and 602 to be laid out.
FIG. 11 is a diagram showing layout results 610 and 620 to be evaluated.
FIG. 12 is a diagram showing attention areas of layout results 610 and 620;
FIG. 13 is a functional block diagram showing a configuration of a layout evaluation apparatus according to the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 100 ... Evaluation part, 110 ... Layout result storage part, 115 ... Attention area extraction part, 120 ... Feature-value calculation part, 130 ... Attention area distribution model storage part, 140, 150 ... Layout result evaluation part, 200 ... Learning part, 210 ... layout result designation input unit, 215 ... attention area extraction unit, 220 ... feature quantity calculation unit, 230 ... feature learning unit, 300 ... neural network, 600, 602 ... image, 610, 620 ... layout result

Claims (14)

A layout evaluation system for evaluating a layout,
Attention area extraction means for calculating a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracting the attention area of the layout result based on the calculated degree of attraction;
Feature quantity calculating means for calculating a feature quantity related to a visual guidance field for the attention area extracted by the attention area extracting means;
Layout result evaluation means for evaluating the layout result based on the feature quantity calculated by the feature quantity calculation means;
Layout evaluation model storage means for storing a layout evaluation model in which a relationship between a feature amount relating to a visual guidance field and an evaluation of the layout result is learned;
The feature amount calculation means calculates the strength of the visual guidance field for the attention region extracted by the attention region extraction means, and calculates the guidance field feature amount indicating the calculated visual guidance field strength as the feature amount. And
The layout result evaluation unit evaluates the layout result based on the guidance field feature amount calculated by the feature amount calculation unit and the layout evaluation model of the layout evaluation model storage unit. A layout evaluation system for evaluating a layout,
Attention area extraction means for calculating a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracting the attention area of the layout result based on the calculated degree of attraction;
Feature quantity calculating means for calculating a feature quantity related to a visual guidance field for the attention area extracted by the attention area extracting means;
Layout result evaluation means for evaluating the layout result based on the feature quantity calculated by the feature quantity calculation means;
Layout evaluation model storage means for storing a layout evaluation model in which a relationship between a feature amount relating to a visual guidance field and an evaluation of the layout result is learned;
The feature amount calculating means calculates a visual guidance field energy for the attention area extracted by the attention area extraction means, calculates an energy feature amount indicating the calculated visual guidance field energy as the feature quantity,
The layout result evaluation unit evaluates the layout result based on an energy feature amount calculated by the feature amount calculation unit and a layout evaluation model of the layout evaluation model storage unit. In any one of Claim 1 and 2,
The layout evaluation model is a neural network that has learned a relationship between a feature amount relating to a visual induction field and an evaluation of the layout result,
The layout result evaluation unit inputs the feature amount calculated by the feature amount calculation unit to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. In any one of Claims 1 thru | or 3,
A feature that calculates a visual guidance field for the attention area extracted by the attention area extraction unit, obtains an equipotential line from the calculated visual guidance field, and calculates a complexity feature amount indicating the complexity of the equipotential line A quantity calculating means;
Layout evaluation model storage means for storing a layout evaluation model that learns the relationship between the complexity feature amount indicating the complexity of the equipotential line and the evaluation of the layout result;
The feature amount calculating means calculates a visual guidance field for the attention area extracted by the attention area extraction means, obtains an equipotential line from the calculated visual guidance field, and indicates a complexity of the equipotential line Degree feature amount,
The layout result evaluation unit evaluates the layout result based on the complexity feature amount calculated by the feature amount calculation unit and the layout evaluation model of the layout evaluation model storage unit. In claim 4,
The layout evaluation model is a neural network that has learned the relationship between the complexity feature amount indicating the complexity of the equipotential line and the evaluation of the layout result,
The layout result evaluation unit inputs the complexity feature amount calculated by the feature amount calculation unit to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. . A layout evaluation system for evaluating a layout,
Attention area extraction means for calculating a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracting the attention area of the layout result based on the calculated degree of attraction;
Attention area extraction means induction field characteristics Ryo及 beauty shows induction field strength of the visual for the region of interest extracted by, and distribution of the region of interest defined by the energy characteristic quantity indicative of the energy of the visual field of induction the Layout evaluation model storage means for storing a layout evaluation model that has learned a relationship with evaluation of a layout result;
Layout result evaluation means for evaluating the layout result based on the distribution of the region of interest and the layout evaluation model of the layout evaluation model storage means;
A layout evaluation system comprising: In claim 6,
The layout evaluation model is a neural network that has learned the relationship between the distribution of the region of interest and the evaluation of the layout result,
The layout result evaluation means inputs a distribution of the attention area extracted by the attention area extraction means to the neural network, and outputs an output value of the neural network as an evaluation of the layout result. . In any one of Claim 6 and 7,
When the plurality of attention areas are extracted by the attention area extraction section, the layout result evaluation means evaluates the layout result based on a distribution of the upper predetermined number having a high degree of attraction among the plurality of attention areas. A layout evaluation system characterized by A layout evaluation program for evaluating a layout,
For a computer including an attention area extraction unit, a feature amount calculation unit, and a layout result evaluation unit,
The attention area extraction means calculates a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracts a attention area of the layout result based on the calculated degree of attraction;
A feature amount calculating step in which the feature amount calculating means calculates a feature amount relating to a visual guidance field for the attention area extracted in the attention area extraction step;
A layout result evaluation step in which the layout result evaluation means evaluates the layout result based on the feature amount calculated in the feature amount calculation step;
The layout evaluation model storage means includes a program for executing a process including a layout evaluation model storage step for storing a layout evaluation model in which a relationship between a feature amount relating to a visual guidance field and an evaluation of the layout result is learned,
The feature amount calculating step calculates a visual guidance field strength for the attention region extracted in the attention region extraction step, and calculates a guidance field feature amount indicating the calculated visual guidance field strength as the feature amount. And
The layout result evaluation step evaluates the layout result based on the guidance field feature amount calculated in the feature amount calculation step and the layout evaluation model stored in the layout evaluation model storage step. . A layout evaluation program for evaluating a layout,
For a computer including an attention area extraction unit, a feature amount calculation unit, a layout result evaluation unit, and a layout evaluation model storage unit,
The attention area extraction means calculates a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracts a attention area of the layout result based on the calculated degree of attraction;
A feature amount calculating step in which the feature amount calculating means calculates a feature amount relating to a visual guidance field for the attention area extracted in the attention area extraction step;
A layout result evaluation step in which the layout result evaluation means evaluates the layout result based on the feature amount calculated in the feature amount calculation step;
The layout evaluation model storage means includes a program for executing a process including a layout evaluation model storage step for storing a layout evaluation model in which a relationship between a feature amount relating to a visual guidance field and an evaluation of the layout result is learned,
The feature amount calculating step calculates energy of a visual guidance field for the attention region extracted in the attention region extraction step, calculates an energy feature amount indicating the calculated visual guidance field energy as the feature amount,
The layout evaluation program evaluates the layout result based on the energy feature amount calculated in the feature amount calculation step and the layout evaluation model stored in the layout evaluation model storage step. A layout evaluation program for evaluating a layout,
To a computer comprising attention area extraction means, layout evaluation model storage means for storing a layout evaluation model in which the relationship between the distribution of attention areas in layout results and evaluation of the layout results is learned, and layout result evaluation means And
The attention area extraction means calculates an attention degree that is a parameter that matches human subjectivity based on the layout result, and extracts the attention area based on the calculated degree of attention; and
The layout result evaluating means is an attention defined by the guidance field feature amount indicating the strength of the visual guidance field for the attention region extracted in the attention region extraction step and the energy feature amount indicating the energy of the visual guidance field. A layout evaluation program comprising: a program for executing a process comprising a layout result evaluation step for evaluating the layout result based on an area distribution and a layout evaluation model of the layout evaluation model storage means . A layout evaluation method for evaluating a layout using a layout evaluation system comprising an attention area extraction unit, a feature amount calculation unit, a layout result evaluation unit, and a layout evaluation model storage unit,
The attention area extraction means calculates a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracts a attention area of the layout result based on the calculated degree of attraction;
A feature amount calculating step in which the feature amount calculating means calculates a feature amount relating to a visual guidance field for the attention area extracted in the attention area extraction step;
A layout result evaluation step in which the layout result evaluation means evaluates the layout result based on the feature amount calculated in the feature amount calculation step;
A layout evaluation model storage means for storing a layout evaluation model storing a layout evaluation model in which a relationship between a feature amount related to a visual guidance field and an evaluation of the layout result is stored;
The feature amount calculating step calculates a visual guidance field strength for the attention region extracted in the attention region extraction step, and calculates a guidance field feature amount indicating the calculated visual guidance field strength as the feature amount. And
The layout result evaluation step evaluates the layout result based on the guidance field feature amount calculated in the feature amount calculation step and the layout evaluation model stored in the layout evaluation model storage step. A layout evaluation method for evaluating a layout using a layout evaluation system comprising an attention area extraction unit, a feature amount calculation unit, a layout result evaluation unit, and a layout evaluation model storage unit,
The attention area extraction means calculates a degree of attraction that is a parameter suitable for human subjectivity based on the layout result, and extracts a attention area of the layout result based on the calculated degree of attraction;
A feature amount calculating step in which the feature amount calculating means calculates a feature amount relating to a visual guidance field for the attention area extracted in the attention area extraction step;
A layout result evaluation step in which the layout result evaluation means evaluates the layout result based on the feature amount calculated in the feature amount calculation step;
A layout evaluation model storage means for storing a layout evaluation model storing a layout evaluation model in which a relationship between a feature amount related to a visual guidance field and an evaluation of the layout result is stored;
The feature amount calculating step calculates energy of a visual guidance field for the attention region extracted in the attention region extraction step, calculates an energy feature amount indicating the calculated visual guidance field energy as the feature amount,
The layout result evaluation step evaluates the layout result based on the energy feature quantity calculated in the feature quantity calculation step and the layout evaluation model stored in the layout evaluation model storage step. A layout evaluation system comprising attention area extraction means, layout evaluation model storage means for storing a layout evaluation model in which the relationship between the distribution of attention areas in layout results and the evaluation of the layout results is learned, and layout result evaluation means A layout evaluation method for evaluating a layout using
The attention area extracting means calculates an attention degree that is a parameter suitable for human subjectivity based on the layout result, and extracts the attention area based on the calculated attention degree; and the layout result The distribution of the attention area defined by the guidance field feature quantity indicating the strength of the visual guidance field for the attention area extracted by the evaluation means in the attention area extraction step and the energy feature quantity indicating the energy of the visual guidance field. And a layout result evaluation step for evaluating the layout result based on a layout evaluation model of the layout evaluation model storage means.

Images